Read 1

(2025) A Variable Admittance Control Strategy for Stable and Compliant Human-Robot Physical Interaction, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2024.3519885.
Keywords: Accuracy; Admittance; Convex optimisation; Force; Human-robot interaction; model predictive control; Observers; Optimization; Oscillators; physical human-robot interaction; Probability density function; Robots; Tuning; variable admittance control.
Abstract: Admittance control is an important method for providing collaborative robots with precise manipulation and flexible contact behavior in industrial settings that often involve physical interaction. However, too rigid or high-frequency interactions by non-specialists will jeopardise the stability of the system. To address this issue, this research presents a novel admittance control framework for collaborative robots to detect oscillatory states and maintain stability by adjusting the controller parameters. In particular, a recursive haptic stability observer is designed to provide a quantitative assessment of the system stability, while a variable admittance controller based on model predictive control is constructed for optimal tuning of stability and flexibility to meet the requirements of a variable task. The effectiveness of the present algorithm is verified in experiments simulating two industrial tasks conducted on the AUBO I5 collaborative robot with 23 volunteers. In addition, the algorithm is tested for application in real collaborative tasks.

(2025) A Multi-Task Energy-Aware Impedance Controller for Enhanced Safety in Physical Human-Robot Interaction, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2024.3519871.
Keywords: Collision avoidance; Compliance and impedance control; human-robot collaboration; Human-robot interaction; Impedance; Load flow; Multitasking; Null space; physical human-robot interaction; Regulation; Robots; Safety; safety in HRI; Service robots.
Abstract: In physical human-robot interaction (pHRI), ensuring human safety in all tasks conducted by the robot is crucial. Traditional compliance control strategies, such as admittance and impedance control, often lead to unpredictable robot behavior due to incidents like contact loss or unexpected external forces, which can cause significant harm to humans. To overcome these limitations, this study introduces a multi-task energy-aware impedance controller for kinematically redundant robots. This controller extends the energy-aware impedance control strategy, which ensures the passivity and safety of a single task using a virtual global energy tank, to kinematically redundant robots performing multiple tasks. The proposed controller effectively regulates the power flow of all tasks performed by the robot through a single global energy tank, ensuring the safety and passivity of the tasks. Experimental results in a shared environment, where external forces are simultaneously applied to the end-effector and the third joint of the Franka Emika Panda, showed that the robot’s energy and power, as well as the power of all tasks, consistently remained within predefined thresholds. Additionally, when comparing the proposed controllers with controller that do not consider null space projection in the power regulation stage and controller that do not regulate the robot’s power, our approach effectively managed the robot’s energy and power and the power of all tasks, ensuring passivity and enhanced safety.

(2024) Repetitive Impedance Learning-Based Physically Human-Robot Interactive Control, in IEEE transaction on neural networks and learning systems, DOI: 10.1109/TNNLS.2023.3243091.
Keywords: Algorithms; Dynamics; Electric Impedance; Force; Humans; Impedance; Impedance control; impedance learning; iterative learning; Learning; Machine Learning; Neural Networks, Computer; physical human-robot interaction (PHRI); repetitive learning control; robot control; Robot kinematics; Robot sensing systems; Robotics - methods; Robots; Task analysis.
Abstract: Model-based impedance learning control can provide variable impedance regulation for robots through online impedance learning without interaction force sensing. However, the existing related results only guarantee the closed-loop control systems to be uniformly ultimately bounded (UUB) and require the human impedance profiles being periodic, iteration-dependent, or slowly varying. In this article, a repetitive impedance learning control approach is proposed for physical human-robot interaction (PHRI) in repetitive tasks. The proposed control is composed of a proportional-differential (PD) control term, an adaptive control term, and a repetitive impedance learning term. Differential adaptation with projection modification is designed for estimating robotic parameters uncertainties in the time domain, while fully saturated repetitive learning is proposed for estimating time-varying human impedance uncertainties in the iterative domain. Uniform convergence of tracking errors is guaranteed by the PD control and the use of projection and full saturation in the uncertainties estimation and is theoretically proved based on a Lyapunov-like analysis. In impedance profiles, the stiffness and damping are composed of an iteration-independent term and an iteration- dependent disturbance, which are estimated by repetitive learning and compressed by the PD control, respectively. Therefore, the developed approach can be applied to the PHRI where iteration-dependent disturbances exist in the stiffness and damping. The control effectiveness and advantages are validated by simulations on a parallel robot in a repetitive following task.Model-based impedance learning control can provide variable impedance regulation for robots through online impedance learning without interaction force sensing. However, the existing related results only guarantee the closed-loop control systems to be uniformly ultimately bounded (UUB) and require the human impedance profiles being periodic, iteration-dependent, or slowly varying. In this article, a repetitive impedance learning control approach is proposed for physical human-robot interaction (PHRI) in repetitive tasks. The proposed control is composed of a proportional-differential (PD) control term, an adaptive control term, and a repetitive impedance learning term. Differential adaptation with projection modification is designed for estimating robotic parameters uncertainties in the time domain, while fully saturated repetitive learning is proposed for estimating time-varying human impedance uncertainties in the iterative domain. Uniform convergence of tracking errors is guaranteed by the PD control and the use of projection and full saturation in the uncertainties estimation and is theoretically proved based on a Lyapunov-like analysis. In impedance profiles, the stiffness and damping are composed of an iteration-independent term and an iteration- dependent disturbance, which are estimated by repetitive learning and compressed by the PD control, respectively. Therefore, the developed approach can be applied to the PHRI where iteration-dependent disturbances exist in the stiffness and damping. The control effectiveness and advantages are validated by simulations on a parallel robot in a repetitive following task.;Model-based impedance learning control can provide variable impedance regulation for robots through online impedance learning without interaction force sensing. However, the existing related results only guarantee the closed-loop control systems to be uniformly ultimately bounded (UUB) and require the human impedance profiles being periodic, iteration-dependent, or slowly varying. In this article, a repetitive impedance learning control approach is proposed for physical human-robot interaction (PHRI) in repetitive tasks. The proposed control is composed of a proportional-differential (PD) control term, an adaptive control term, and a repetitive impedance learning term. Differential adaptation with projection modification is designed for estimating robotic parameters uncertainties in the time domain, while fully saturated repetitive learning is proposed for estimating time-varying human impedance uncertainties in the iterative domain. Uniform convergence of tracking errors is guaranteed by the PD control and the use of projection and full saturation in the uncertainties estimation and is theoretically proved based on a Lyapunov-like analysis. In impedance profiles, the stiffness and damping are composed of an iteration-independent term and an iteration- dependent disturbance, which are estimated by repetitive learning and compressed by the PD control, respectively. Therefore, the developed approach can be applied to the PHRI where iteration-dependent disturbances exist in the stiffness and damping. The control effectiveness and advantages are validated by simulations on a parallel robot in a repetitive following task.

(2024) SRL-VIC: A Variable Stiffness-Based Safe Reinforcement Learning for Contact-Rich Robotic Tasks, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2024.3396368.
Keywords: Aerospace electronics; Compliance and impedance control; Conformance testing; Impedance; Recovery; reinforce- ment learning (RL); Reinforcement learning; Robot learning; Robotics and automation; robotics and automation in construction; Safety; Stiffness; Task analysis; Task complexity; Trajectory.
Abstract: Reinforcement learning (RL) has emerged as a promising paradigm in complex and continuous robotic tasks, however, safe exploration has been one of the main challenges, especially in contact-rich manipulation tasks in unstructured environments. Focusing on this issue, we propose SRL-VIC : a model-free s afe RL framework combined with a variable impedance controller ( VIC ). Specifically, safety critic and recovery policy networks are pre-trained where safety critic evaluates the safety of the next action using a risk value before it is executed and the recovery policy suggests a corrective action if the risk value is high. Furthermore, the policies are updated online where the task policy not only achieves the task but also modulates the stiffness parameters to keep a safe and compliant profile. A set of experiments in contact-rich maze tasks demonstrate that our framework outperforms the baselines (without the recovery mechanism and without the VIC), yielding a good trade-off between efficient task accomplishment and safety guarantee. We show our policy trained on simulation can be deployed on a physical robot without fine-tuning, achieving successful task completion with robustness and generalization.

(2024) Bayesian Algorithm-Based Force Profiles Optimization of Hip-Assistive Soft Exosuits Under Variable Walking Speeds, in IEEE transactions on medical robotics and bionics, DOI: 10.1109/TMRB.2024.3408308.
Keywords: Algorithms; Assistive devices; Bayes methods; Bayesian analysis; Bayesian optimization; Energy consumption; Energy costs; Exoskeletons; Gait recognition; Hip; hip assistance; Human in the loop; Legged locomotion; Locomotion; metabolic cost; Metabolism; Optimization; Parameters; Real time; Soft exosuits; Walking.
Abstract: Relevant research highlights humans’ capacity to continuously adapt their walking speed to minimize metabolic energy consumption during overground free walking. Past studies have shown that soft exosuits assisting in hip flexion and extension can reduce metabolic costs and regulate gait parameters during human locomotion. This emphasizes the need to fine-tune hip exosuit parameters to align with walking speed, thereby enhancing metabolic efficiency. This study aims to optimize assistive force parameters of hip exosuits across different walking speeds, providing insights for optimizing force profiles in outdoor walking. We employed a human-in-the-loop approach with Bayesian optimization to determine optimal force profiles for hip assistance. Six subjects performed treadmill walking at four fixed speeds (0.84, 1.16, 1.48, and 1.8 m/s), optimizing control parameters for each speed and establishing a Bayesian experience (BXE) linking walking speed to optimal parameters. Furthermore, we developed a real-time force optimization controller based on the BXE for adjusting the force parameters of assistance. Outdoor walking experiments with the same subjects showed that BXE-optimized profiles significantly reduced metabolic costs compared to fixed profiles. This study underscores the importance of optimizing assistive forces for varying walking speeds in humans.

(2024) Adaptive Compensation Tracking Control for Parallel Robots Actuated by Pneumatic Artificial Muscles With Error Constraints, in IEEE Trans. Ind. Inform., DOI: 10.1109/TII.2023.3280321.
Keywords: Adaptation models; Adaptive control; Artificial muscles; Closed loops; Compensation; Control systems design; Controllers; Dynamic models; Lyapunov methods; Lyapunov techniques; Mathematical models; Muscles; Parallel robots; Parameter uncertainty; pneumatic artificial muscles (PAMs); Robot control; Robots; Safety; Tracking control; Tracking errors; Uncertain systems.
Abstract: As pneumatic artificial muscles (PAMs) are similar to biological muscles in structure and movement mechanisms, parallel robots actuated by PAMs have development prospects in rehabilitation and industry, with advantages such as compliance, high safety, strong bearing capacity, and satisfactory dynamic performance. However, the parameter uncertainties and model complexity related to inherent characteristics of parallel robots actuated by PAMs (e.g., time-varying, coupling, hysteresis, creep, and high nonlinearity), bring challenges to accurate dynamic modeling and controller design. Therefore, to achieve satisfactory tracking performance, this article presents an adaptive compensation tracking controller with error constraints for parallel robots actuated by PAMs. The proposed controller deals with parameter uncertainties by estimating system parameters to ensure accurate tracking, which is indicated as an effective solution for a combination of PAMs and parallel robots. Furthermore, using desired trajectory signals in the complicated regression matrix, the online computational burden is significantly reduced. Moreover, to improve operation safety further, an auxiliary term with a theoretical demonstration guarantees that the tracking errors are maintained within allowable ranges. Then, the closed-loop stability is demonstrated by Lyapunov techniques. As far as we know, it is the first time that the challenges of parameter uncertainties, computational burdens, and error constraints of parallel robots actuated by PAMs are simultaneously addressed, which has both theoretical significance and practical value. Finally, the hardware experiments are implemented under different scenarios, and the results indicate that the proposed method achieves satisfactory tracking performance.

(2024) Hierarchical Human Motion Intention Prediction for Increasing Efficacy of Human-Robot Collaboration, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2024.3430131.
Keywords: Collaboration; Collaborative robots; Decoding; efficacy; human motion decoding; human motion intent; human motion prediction; Human-robot collaboration; Human-robot interaction; Predictive models; Robot motion; Task analysis; Trajectory.
Abstract: When humans and robots work together to accomplish tasks with dynamic uncertainty, the robots should perceive human motion intentions so as to cooperate with humans and increase efficacy. In this study, we propose a hierarchical motion intention prediction model for human-robot collaboration, in which the bottom level acquires human motion information, the middle level recognizes motion states and the high level predicts motion intentions. Compared with existing methods, our model fuses task-level human behavioral pattern prediction with instantaneous continuous motion intent decoding. Therefore, the robot controller can generate a collaborative trajectory in advance and adjust the key parameters (forces and velocities, etc.) in real time according to human motions. We quantitatively verify the proposed model with 10 subjects in the human-robot sawing task. The results show that the hierarchical model can effectively reduce human energy consumption and improve the average speed of the task. Meanwhile, subjective metrics indicate that subjects believe robots employing hierarchical models as capable of fostering improved cooperation and delivering greater assistance. Our study systematically proves that the proposed hierarchical model significantly enhanced the efficiency of human-robot co-manipulation, marking a step forward compared with existing works. Future studies will be focused on investigating more complex and general tasks.

(2024) Polymorphic Control Framework for Automated and Individualized Robot-Assisted Rehabilitation, in IEEE Trans. Robot., DOI: 10.1109/TRO.2023.3335666.
Keywords: Algorithms; Assistance-as-needed; biocooperative control; Biomechanics; control framework; Control methods; Control systems; Controllers; Decision making; Design; Invariants; Medical treatment; Modular design; on-demand supervision; polymorphic control; rehabilitation robotics; Rehabilitation robots; Robot sensing systems; Robots; Task analysis; Training.
Abstract: Robots were introduced in the field of upper limb neurorehabilitation to relieve the therapist from physical labor, and to provide high-intensity therapy to the patient. A variety of control methods were developed that incorporate patients’ physiological and biomechanical states to adapt the provided assistance automatically. Higher level states, such as selected type of assistance, chosen task characteristics, defined session goals, and given patient impairments, are often neglected or modeled into tight requirements, low-dimensional study designs, and narrow inclusion criteria so that presented solutions cannot be transferred to other tasks, robotic devices or target groups. In this work, we present the design of a modular high-level control framework based on invariant states covering all decision layers in therapy. We verified the functionality of our framework on the assistance and task layer by outlaying the invariant states based on the characteristics of 20 examined state-of-the-art controllers. Then, we integrated four controllers on each layer and designed two algorithms that automatically selected suitable controllers. The framework was deployed on an arm rehabilitation robot and tested on one participant acting as a patient. We observed plausible system reactions to external changes by a second operator representing a therapist. We believe that this work will boost the development of novel controllers and selection algorithms in cooperative decision-making on layers other than assistance, and eases transferability and integration of existing solutions on lower layers into arbitrary robotic systems.

(2024) A Hybrid Adaptive Controller for Soft Robot Interchangeability, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2023.3337705.
Keywords: Actuation; Adaptive control; adaptive controller; Artificial neural networks; Bionics; Controllers; hybrid controller; Hybrid power systems; Jacobian matrices; Kinematics; Long short term memory; LSTM; Molds; Neural networks; Recurrent neural networks; Robot dynamics; Robot kinematics; Robot sensing systems; Robots; Soft robot control; Soft robotics; Softness; Trajectory.
Abstract: Soft robots have been leveraged in considerable areas like surgery, rehabilitation, and bionics due to their softness, flexibility, and safety. However, it is challenging to produce two same soft robots even with the same mold and manufacturing process owing to the complexity of soft materials. Meanwhile, widespread usage of a system requires the ability to replace inner components without highly affecting system performance, which is interchangeability. Due to the necessity of this property, a hybrid adaptive controller is introduced to achieve interchangeability from the perspective of control approaches. This method utilizes an offline-trained recurrent neural network controller to cope with the nonlinear and delayed response from soft robots. Furthermore, an online optimizing kinematics controller is applied to decrease the error caused by the above neural network controller. Soft pneumatic robots with different deformation properties but the same mold have been included for validation experiments. In the experiments, the systems with different actuation configurations and the different robots follow the desired trajectory with errors of [Formula Omitted] and [Formula Omitted] compared with the working space length, respectively. Such an adaptive controller also shows good performance on different control frequencies and desired velocities. This controller is also compared with a model-based controller in simulation. This controller endows soft robots with the potential for wide application, and future work may include different offline and online controllers. A weight parameter adjusting strategy may also be proposed in the future.;Soft robots have been leveraged in considerable areas like surgery, rehabilitation, and bionics due to their softness, flexibility, and safety. However, it is challenging to produce two same soft robots even with the same mold and manufacturing process owing to the complexity of soft materials. Meanwhile, widespread usage of a system requires the ability to replace inner components without highly affecting system performance, which is interchangeability. Due to the necessity of this property, a hybrid adaptive controller is introduced to achieve interchangeability from the perspective of control approaches. This method utilizes an offline-trained recurrent neural network controller to cope with the nonlinear and delayed response from soft robots. Furthermore, an online optimizing kinematics controller is applied to decrease the error caused by the above neural network controller. Soft pneumatic robots with different deformation properties but the same mold have been included for validation experiments. In the experiments, the systems with different actuation configurations and the different robots follow the desired trajectory with errors of \mathbf{3.3\pm 2.9\%} and \mathbf{4.3\pm 4.1\%} compared with the working space length, respectively. Such an adaptive controller also shows good performance on different control frequencies and desired velocities. This controller is also compared with a model-based controller in simulation. This controller endows soft robots with the potential for wide application, and future work may include different offline and online controllers. A weight parameter adjusting strategy may also be proposed in the future.

(2024) A Human-Robot Collaboration Controller Utilizing Confidence for Disagreement Adjustment, in IEEE Trans. Robot., DOI: 10.1109/TRO.2024.3370025.
Keywords: Adaptation models; Adaptive control; Collaboration; Confidence intervals; Controllers; Cooperation; Dynamic models; Estimation; Human motion; Impedance; intention estimation; Neural networks; Physical human-robot interaction; Predictive models; reinforcement learning; Robot dynamics; Robots; Task analysis.
Abstract: With the development of collaborative robots, the demand for efficient and safe physical human-robot interaction (pHRI) is significantly increasing. In this paper, a two-loop pHRI controller is proposed to reduce the disagreement in human-robot cooperation and to enhance the level of robot assistance. In the outer loop, a human motion intention estimator is designed, combining the strength of model-free and model-based approaches. It estimates the human’s desired movement position and provides the confidence level for the estimated value. Subsequently, the estimated value is tracked by the inner loop controller, and the confidence level is used to adjust the robot’s behavior in order to reduce the disagreement. In the inner loop, a neuro-adaptive controller with a variable reference model is designed to achieve the efficient pHRI. A neural network is applied to compensate for the nonlinearity of the robot dynamics, gradually aligning the input-output characteristic of the robot dynamic model with the one of the reference model. To minimize the human-robot disagreement during the collaboration process and to enhance the robot’s assistance level, a reinforcement learning method is proposed to adjust parameters of the reference model. The proposed control scheme is implemented on a Franka Panda robot and validated through the point-to-point movement simulation and a real-world human-robot lifting experiment. Results suggest that compared to other methods, the proposed approach can indeed reduce the human-robot disagreement and improve the robot assistance level.;With the development of collaborative robots, the demand for efficient and safe physical human–robot interaction (pHRI) is significantly increasing. In this article, a two-loop pHRI controller is proposed to reduce the disagreement in human–robot cooperation and to enhance the level of robot assistance. In the outer loop, a human motion intention estimator is designed, combining the strength of model-free and model-based approaches. It estimates the human’s desired movement position and provides the confidence level for the estimated value. Subsequently, the estimated value is tracked by the inner loop controller, and the confidence level is used to adjust the robot’s behavior in order to reduce the disagreement. In the inner loop, a neuro-adaptive controller with a variable reference model is designed to achieve the efficient pHRI. A neural network is applied to compensate for the nonlinearity of the robot dynamics, gradually aligning the input–output characteristic of the robot dynamic model with the one of the reference model. To minimize the human–robot disagreement during the collaboration process and to enhance the robot’s assistance level, a reinforcement learning method is proposed to adjust parameters of the reference model. The proposed control scheme is implemented on a Franka Panda robot and validated through the point-to-point movement simulation and a real-world human–robot lifting experiment. Results suggest that compared to other methods, the proposed approach can indeed reduce the human–robot disagreement and improve the robot assistance level.

(2024) Saturated Sliding Mode Control Scheme for a New Wearable Back-Support Exoskeleton, in IEEE Trans. Autom. Sci. Eng., DOI: 10.1109/TASE.2023.3241619.
Keywords: Back pain; Control stability; Control systems design; Controllers; Ergonomics; Exoskeletons; Fasteners; Force; linear extended state observer; Lumbar spinal rehabilitation exoskeleton; Motion control; Myoelectricity; Observers; Pain; Patient rehabilitation; Push rods; Rehabilitation; Rehabilitation robots; Robot control; Robot dynamics; Robots; saturated control; Sliding mode control; sliding mode observer; Spinal cord injury; State observers; Tracking errors; Trajectories.
Abstract: Low back pain has been torturing people around the world as a common and chronic disease, the main inducement of which is lumbar disc herniation (LDH). To alleviate patients’ pain noninvasively, this paper proposes a new lumbar spinal rehabilitation exoskeleton. This exoskeleton is composed of two bands connected by four motor-driven piston pushrods, where the range of band can be adjusted to adapt to people with different waistlines. The four pushrods provide support forces to hold the upper body to relieve the burden of lumbar. The joints between pushrods and band are universal pairs and spherical pairs. The whole structure can be regarded as a parallel robot, thus the supporting and waist tracking performance can be determined by motion control effect of pushrods. Considering the motor torque of pushrods is limited, in this paper, a saturated sliding mode control scheme is proposed. Meanwhile, an extended state observer (ESO) is employed to estimate the external disturbance and internal uncertainties, then the estimates of perturbations will be compensated by the feedback scheme. Furthermore, to enhance the disturbance-tracking performance of the ESO an integrated sliding mode observer is proposed to compensate the tracking errors of ESO. The stabilities of controller and observers are proved by Lyapunov stability theory. Finally, a simulation and two experiments are conducted to verify the performance of the proposed controller and the new exoskeleton. The simulation results show that the novel/new controller can drive the new exoskeleton to move along with the desirable trajectory to support the upper body so that alleviating the burden of waist. The results of myoelectricity experiments also show favorable effectiveness of exoskeleton on supporting upper body in different postures. Note to Practitioners—this article was inspired by the concerns of low back pain that many people are suffering. Most existing rehabilitation robots tend to focus on supporting lower body and arms, while back-support rehabilitation robots are barely seen. Although we can see a few upper-body supporting robots in sporadic papers, these studies have not proposed a good control scheme yet. Besides, it is difficult to find a rehabilitation robot with simple structure as well as conforming to ergonomics. To resolve the shortcomings of the structure of existing robots and control models, we proposed a completely new rehabilitation robot with novel mechanical structure. Meanwhile, a new saturated sliding mode and novel integrated observer are employed in the position-force control system to enhance the stability and controllability of the whole robot control system. Then, some elaborately designed experiments are conducted to test the proposed robot and the control system. Numerical results demonstrate that the proposed method can significantly increase the trajectory tracking performance and alleviate low back pain.

(2024) Progressive Learning based Assist-as-Needed Control for Ankle Rehabilitation, in IEEE Trans. Cogn. Dev. Syst., DOI: 10.1109/TCDS.2024.3455795.
Keywords: Ankle; Force; Fuzzy logic; Iterative methods; Measurement; Performance-based approach; Progressive learning; Rehabilitation robotics; Robots; Torque; Training.
Abstract: This paper proposes a progressive learning based assist-as-needed (AAN) control scheme for ankle rehabilitation. To quantify the training performance, a fuzzy logic (FL) system is established to generate a holistic metric based on multiple kinematic and dynamic indicators. Subsequently, a cost function that contains both the tracking error and robot stiffness is constructed. A novel learning scheme is then proposed to enhance subjects’ engagement, leveraging the FL metric to uphold a declining trend in the robot’s stiffness. The system stability is analyzed using the Lyapunov theory, the control ultimate bounds are specified and the effects of parameter tuning are discussed. Experiments are conducted on an ankle robot and the minimal assist-as-needed (MAAN) scheme is adopted for comparison. With a training session consisting of 11 trials, the quantitative performance evaluations, individual error convergences, progressive stiffness learning and human-robot interaction are evaluated. It is shown that within 8 trials under the progressive AAN and MAAN, the robot assistive torques have an average reduction of 13.45% and 20.25% while subjects’ active torques are increased by 56.53% and 58.39%, respectively. During the late stage of training, the progressive AAN further improves two criteria by 9.44% and 6.29%, while the MAAN partially loses subjects’ participation (active torques are reduced by 36.38%) due to the occurrence of motion adaption.

(2024) Force Tracking Control With Adaptive Stiffness and Iterative Position of Hip-Assistive Soft Exosuits, in IEEE Trans. Autom. Sci. Eng., DOI: 10.1109/TASE.2023.3339779.
Keywords: Actuators; Adaptation models; adaptive stiffness control; Force; Force control; force tracking; Hip; hip assistance; Iterative methods; iterative optimization; Legged locomotion; Load modeling; Soft electronics; Soft exosuits.
Abstract: Soft exosuits feature nonlinear, low stiffness, and hysteretic behavior, presenting different mechanical responses during loading and unloading of assistive force. Delivering the desired force to the wearer is a challenge in such a system. To address this issue, this article proposed a novel control strategy with adaptive stiffness and iterative position to optimize the assistive forces of the actuators in compliance with the humanexosuit interface stiffness and relative motion. Specifically, a stiffness model was proposed to describe the force loading and unloading behaviors, further an adaptive stiffness-based force controller was designed to improve the force tracking for whole profile by adaptively adjusting the stiffness parameters related to loading and unloading, and iteratively compensating the position error associated with the force amplitude in the model on a step-by-step basis. This control strategy was implemented on a soft exosuit for hip flexion and extension assistance, and its performance was evaluated through walking tests on six subjects. The results showed that after 4 or 5 iterations of optimization based on the initial parameter settings, the proposed controller could comply with the human-exosuit interface stiffness and achieved an improved force tracking, with a minimal root-mean-square error of 7.5 N in desired force profiles tracking, and a metabolic gain of 14.8%, demonstrating a promising potential of the proposed controller for improving the force tracking and the economy of human walking. Note to Practitioners-Soft exosuits have shown promising outcomes in augmenting human walking and reducing fatigue. However, delivering the desired assistive force accurately to the wearer remains challenging due to the nonlinear and variable stiffness nature of the human-exosuit interaction during walking. This article proposed a control strategy for improving the force tracking performance by optimizing the force-positional relationship of the actuators. Specifically, a stiffness model was established to define the relationship between the force and position of the actuators. This model acts as a virtual spring between the wearer and the exosuit, and stiffness of the spring was adjusted to adapt the wearer. On the Basis of this model, we developed a force tracking controller with adaptive stiffness. It can transmit accurately a desired force trajectory to the wearer by adaptively optimizing the stiffness parameters and iteratively compensating the position error in stiffness model. This controller was implemented on a soft exosuit for hip flexion and extension assistance. The results of treadmill and outdoor walking tests with six subjects confirm the significant effect on optimizing the force-positional relationship, improving the force tracking performance, as well as avoiding force hystereses or friction losses. This research addresses the challenge of force control in human-exosuit interaction with variable stiffness characteristics. The limitation of this work is that it only focuses on hip joint assistance. In the future, we hope to extend this control strategy to other joints to achieve the generality of the approach.

(2024) Dynamic Collaborative Workspace Based on Human Interference Estimation for Safe and Productive Human-Robot Collaboration, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2024.3405352.
Keywords: Safety; Collaboration; Productivity; Interference; Collision avoidance; Monitoring; Collaborative robots; Human-robot interaction; Dynamic collaborative workspace; human-robot collaboration; human interference; safety and productivity.
Abstract: Collaborative robots that operate safely close to workers without fences have attracted attention, but few examples of such human-robot collaboration (HRC) have been seen in factories. The main reason is the difficulty in balancing safety and productivity. Current fenceless HRC systems stop the robot when a human enters the collaborative workspace $( {\mathscr{C}} )$ where both human and robot can work to ensure safety, which ISO/TS15066 regulates. The robot stops even when the human is far enough away, so productivity is drastically decreased (FCW, Fixed ${\mathscr{C}}$). If a system could identify the human-work area, designate it as a no-entry space in ${\mathscr{C}}$ for the robot $( }^{\bm{P}}}} )$, and dynamically set the closed ${\mathscr{C}}$ $( }^{\bm{C}}}} )$ with shrinking ${\mathscr{C}}$ by $}^{\bm{P}}}$, productivity would improve thanks to enabling the robot to work in $}^{\bm{C}}}$ and safety would be ensured thanks to allowing the human to continue working in $}^{\bm{P}}}$. In this study, we propose a new concept of a dynamic collaborative workspace (DCW) that dynamically sets $}^{\bm{C}}}$ and $}^{\bm{P}}}$ based on the human’s predicted trajectory. It also provides visual and auditory prompts to enable the human to understand DCW states, i.e., when a human enters ${\mathscr{C}}$, ${\mathscr{C}}$ is changed, and the robot is in emergency mode. We compared four HRC systems using a real robot arm: two conventional FCW ones with and without fences and two proposed DCW ones with and without a state indicator and found that the proposed system with a state indicator has the best productivity and ensures the same level of safety as the conventional system with fences.

(2024) A Knowledge Transfer-Based Personalized Human-Robot Interaction Control Method for Lower Limb Exoskeletons, in IEEE Sens. J., DOI: 10.1109/JSEN.2024.3479239.
Keywords: Accuracy; Control methods; Customization; Effectiveness; Exoskeleton; Exoskeletons; Feature extraction; Human engineering; Human motion; Human-robot interaction; Intent recognition; Knowledge management; Knowledge transfer; Legged locomotion; Modules; Muscles; Noise prediction; personalized intent recognition; Robot control; Robots; Sensors; surface electromyography (sEMG); Transfer learning; Transformers.
Abstract: Accurate intent recognition by patients while wearing exoskeletons is crucial during their rehabilitation exercises. In this article, a transfer learning framework for human-robot interaction (EMGTnet-KTD) is proposed to predict human movement intentions in human-robot interactions through surface electromyography (sEMG) signals. EMGTnet-KTD consists of a pretrained EMGTnet model and a knowledge transfer module. First, EMGTnet is designed based on a Transformer network. A temporal and spatial domain feature fusion module has been introduced on top of the Transformer network, and the inputs have been reconfigured to enable it to utilize the relationship between before and after human actions. In addition, the knowledge transfer module is composed of a feature extraction layer, a noise reduction layer, and the personalized human lower limb dynamics controller. To evaluate the effectiveness of the proposed method, an experimental validation of our self-collected dataset from seven subjects is performed. The results show that our method achieves better results than other continuous motion prediction methods. Finally, to validate that the generation angle conforms to human physiology, walking experiments involving the use of an exoskeleton are conducted. The experiments demonstrate the effectiveness of the proposed framework and its implementability for exoskeletons.

(2024) Composite Learning Variable Impedance Robot Control With Stability and Passivity Guarantees, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2023.3327675.
Keywords: Adaptation models; Adaptive control; Closed loops; Collaboration; compliance and impedance control; Control stability; Damping; Dynamic stability; Effectiveness; Excitation; Feedback control; Human engineering; Human-robot interaction; Impedance; Learning; Manufacturing processes; parameter convergence; Parameter estimation; Parameter identification; physical human-robot interaction; Robot control; Robot dynamics; Robot kinematics; robot safety; Robots; Safety; System identification; Uncertainty; Weighting functions.
Abstract: Variable impedance control (VIC) is paramount for robots to improve safety and effectiveness in physical human-robot interaction. However, achieving variable target impedance with guaranteed stability is not trivial, particularly under parametric uncertainty in the robot dynamics. This letter proposes a composite learning-based VIC (CL-VIC) strategy with three control modes, including robot-dominant, human-dominant, and collaboration modes, to achieve target impedance under parametric uncertainty. A multi-mode adaptive control scheme is defined by well-designed weighting functions to ensure smooth mode transitions. A composite learning mechanism is employed for exact parameter estimation such that target impedance models can be achieved under interval excitation that is much weaker than persistent excitation, where the latter is a sufficient condition for exact parameter estimation in classical system identification. The proposed method can achieve variable active stiffness and damping with guaranteed exponential stability and passivity of the closed-loop system, which ensures safe interaction. Experiments on a collaborative robot with seven degrees of freedom named Franka Emika Panda have validated the effectiveness of the proposed CL-VIC method.

(2024) Whole-Body Intuitive Physical Human-Robot Interaction With Flexible Robots Using Non-Collocated Proprioceptive Sensing, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2024.3354617.
Keywords: Actuators; Compliance and impedance control; End effectors; flexible robotics; Human engineering; Jacobi matrix method; Jacobian matrix; Manipulator dynamics; Manipulators; physical human-robot interaction; Rigid structures; Robot kinematics; Robot sensing systems; Robots; Sensors.
Abstract: This letter presents a method enabling intuitive physical human-robot interaction (pHRI) with flexible robots using an end-point sensing device. The device is a passive serial chain of encoders and lightweight links, mounted in parallel with the manipulator. By measuring the deflection of the end-effector relative to the base, the whole body of the manipulator becomes a potential interaction interface, whether the compliance stems from the links or the joints. The proposed control scheme is a simple joint velocity control that only requires knowledge of the rigid body Jacobian matrix of the manipulator. The approach is validated both in simulation on a simplified model and experimentally on a physical 3-DoF flexible-link flexible-joint serial robot. The results indicate that intuitive pHRI is achieved, with interaction forces under 25 N even for tasks with high dynamics.

(2024) Physical Human-Robot Interaction Control of an Upper Limb Exoskeleton With a Decentralized Neuroadaptive Control Scheme, in IEEE Trans. Control Syst. Technol., DOI: 10.1109/TCST.2023.3338112.
Keywords: Actuators; Adaptation models; Adaptive control; Adaptive decentralized control; Asymptotic stability; Control systems design; Control theory; Controllers; Decentralized control; Degrees of freedom; Exoskeletons; input and state constraints; Liapunov functions; Neural networks; Operators (mathematics); physical human–robot interactions (pHRIs); Power flow; Proportional derivative; Proportional integral derivative; Radial basis function; Rigid structures; Robot control; Robots; Robust control; Safety; Subsystems; Uncertainty; wearable robots.
Abstract: Within the concept of physical human–robot interaction (pHRI), the paramount criterion is the safety of the human operator interacting with a high-degree-of-freedom (DoF) robot. Consequently, there is a substantial demand for a robust control scheme to establish safe pHRI and stabilize nonlinear, high DoF systems. In this article, an adaptive decentralized control strategy is designed to accomplish the abovementioned objectives. A human upper limb model and an exoskeleton model are decentralized and augmented at the subsystem level to enable decentralized control action design. Human exogenous torque (HET), which can resist exoskeleton motion, is estimated using radial basis function neural networks (RBFNNs). Estimating human upper limb and robot rigid body parameters, as well as HET, makes the controller adaptable to different operators, ensuring their physical safety during the interaction. To guarantee both safe operation and stability, the barrier Lyapunov function (BLF) is utilized to adjust the control law. This study also considers unknown actuator uncertainties and constraints to ensure a smooth and secure pHRI. In addition, it is shown that incorporating RBFNNs and BLF into the original virtual decomposition control (VDC) improved its performance. The asymptotic stability of the entire system is established through the concept of virtual stability and virtual power flows (VPFs) under the proposed robust controller. Experimental results are presented and compared with those obtained using proportional-derivative (PD) and proportional-integral-derivative (PID) controllers to showcase the robustness and superior performance of the designed controller, particularly in controlling the last two joints of the robot.

(2024) Variable Admittance Control Using Velocity-Curvature Patterns to Enhance Physical Human-Robot Interaction, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2024.3387110.
Keywords: Admittance control; Curvature; Education; Electrical impedance; Estimation; Force; Human engineering; Human factors; Human factors and human-in-the-loop; Human in the loop; Human motion; Human-robot interaction; intention recognition; Mathematical analysis; physical human-robot interaction; Robot kinematics; Robots; Smoothness; Torque sensors (robotics); Trajectory; Velocity measurement; velocity-curvature patterns.
Abstract: This letter introduces a variable admittance control approach aimed at enhancing intuitive human-robot interaction by considering both direct and indirect human intentions. The magnitude of force serves as a representation of direct intentions, delineating preferences for rapid or precise motions. Drawing from the movement sciences and motor control field, the minimum-jerk model is employed to mirror human motor system control policies and movement behaviors. From this model, velocity-curvature patterns are derived, enabling an intuitive estimation of indirect intentions indicating long-term objectives like trajectory and turning direction. We propose an innovative guidance method, rooted in the estimation of indirect human intentions, enabling the robot to concurrently follow and guide the operator. To assess the efficacy of this approach, both offline simulations and real-time human experiments are conducted on a six-DOF robot and a force/torque sensor. Our comprehensive experiments demonstrate substantial enhancements in both accuracy ([Formula Omitted]) and smoothness ([Formula Omitted]) over fixed admittance control and state-of-the-art variable admittance methods.;This letter introduces a variable admittance control approach aimed at enhancing intuitive human-robot interaction by considering both direct and indirect human intentions. The magnitude of force serves as a representation of direct intentions, delineating preferences for rapid or precise motions. Drawing from the movement sciences and motor control field, the minimum-jerk model is employed to mirror human motor system control policies and movement behaviors. From this model, velocity-curvature patterns are derived, enabling an intuitive estimation of indirect intentions indicating long-term objectives like trajectory and turning direction. We propose an innovative guidance method, rooted in the estimation of indirect human intentions, enabling the robot to concurrently follow and guide the operator. To assess the efficacy of this approach, both offline simulations and real-time human experiments are conducted on a six-DOF robot and a force/torque sensor. Our comprehensive experiments demonstrate substantial enhancements in both accuracy ({\geq\! ext{12.1}{\%}}) and smoothness ({\geq\! ext{18.3}{\%}}) over fixed admittance control and state-of-the-art variable admittance methods.

(2024) Safety-Guaranteed and Task-Consistent Human-Robot Interaction Using High-Order Time-Varying Control Barrier Functions and Quadratic Programs, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2023.3333166.
Keywords: Collision avoidance; Consistency; Control barrier function; Human engineering; human-robot collaboration; Human-robot interaction; quadratic program; Quadratic programming; Robot dynamics; Robot kinematics; Robots; Safety; task consistency; Time varying control; Trajectory.
Abstract: Close human-robot interaction enables the combination of complementary abilities of humans and robots, thereby promoting efficient manufacturing. Human safety is an important aspect of human-robot interaction. To this end, the robot executes an evasive motion for collision avoidance when the human approaches. However, the evasive motion may be inconsistent with the robot’s task resulting in unresumable task failure. In this letter, a control framework is proposed to achieve guaranteed human safety and hierarchical task consistency. First, the high-order time-varying control barrier function (HO-TV-CBF) is proposed to keep a safe distance between human and robot, thereby guaranteeing human safety. Next, to achieve hierarchical task consistency, a hard constraint and a soft constraint are defined systematically. The hard constraint ensures primary task consistency that keeps the task resumable, while the soft constraint together with the hard constraint ensures full task consistency. Finally, two quadratic programs (QPs) are employed to coordinate different control objectives, i.e., human safety and primary task consistency are always guaranteed while full task consistency is ensured whenever possible. Experiments are conducted to validate the proposed control framework with comparisons to existing methods.

(2024) Soft Ankle-Foot Exoskeleton for Rehabilitation: A Systematic Review of Actuation, Sensing, Mechanical Design, and Control Strategy, in IEEE transactions on medical robotics and bionics, DOI: 10.1109/TMRB.2024.3385798.
Keywords: Actuation; Actuators; Control methods; control strategy; Control systems; Exoskeletons; Feet; Human in the loop; Human-robot interaction; mechanical design; Patient rehabilitation; Rehabilitation; Robot control; Soft ankle-foot exoskeleton; Soft robotics; Systematic review; Wearable sensors; Wearable technology.
Abstract: Robot-assisted rehabilitation therapy has become a mainstream trend for the treatment of stroke patients. It can not only relieve physiotherapists from heavy physical duties, but also provide patients with effective ankle-foot rehabilitation and walking assistance. Soft ankle-foot exoskeletons have rapidly evolved in the last decade. This article presents a compressive review of soft ankle-foot exoskeletons in terms of robot actuation, wearable sensor, mechanical design, and control strategy. Representative commercial and laboratory ankle-foot exoskeletons are demonstrated. Special attention is paid to the emerging soft actuators, wearable sensing techniques, and human-in-the-loop and hierarchical control methods. Finally, essential challenges and possible future directions are also analyzed and highlighted in this paper, which can provide reliable guidance on the development of next-generation soft ankle-foot exoskeletons.

(2024) A Physical Human-Robot Interaction Framework for Trajectory Adaptation Based on Human Motion Prediction and Adaptive Impedance Control, in IEEE Trans. Autom. Sci. Eng., DOI: 10.1109/TASE.2024.3415650.
Keywords: adaptive impedance control; Force; human motion prediction; Impedance; Physical human-robot interaction (pHRI); Robots; Speech recognition; Task analysis; Tracking; Trajectory; trajectory adaptation.
Abstract: Physical human-robot interaction (pHRI) plays an important role in robotic. In order for a human operator to be able to easily adapt to interact with a robot, a minimal interaction force in pHRI should be achieved. In this paper, a pHRI framework is proposed to allow the robot to regulate its trajectory adaptively for minimizing the interaction force with small position-tracking errors. The trajectory of the robot is first adjusted by the interaction force which is updated by the performance evaluation index. Then, the human hand motion is predicted based on the autoregressive (AR) model to further adapt the trajectory. Thirdly, an adaptive impedance control method is developed to update the stiffness in the robot impedance controller using surface electromyography (sEMG) signals for robot compliant interaction with the environment. This method allows the human operator to interact with the robot by the interaction force, the hand motion and muscle contraction. By investigating the performance of the proposed method, the interaction force is decreased and a good position tracking accuracy is achieved. Comparative experiments demonstrate the enhanced performance of the proposed method. Note to Practitioners -This paper focuses on developing a novel method that can allow the robot to compliantly interact with the human operator while simultaneously taking into account the trajectory-tracking accuracy and the interaction force in pHRI scenarios. The proposed method has a large application potential in a variety of pHRI tasks, such as human-robot collaborative transporting, curing, assembly, cutting, and so on. In addition, the proposed method can allow the human operator to physically interact with the robot in an easier and more intuitive manner, by taking advantage of human motion prediction and adaptive impedance control. Therefore, it is also potentially utilized for rehabilitation and assistive robots, and robot learning skills from human physical demonstration.

(2024) Intuitive Human-Robot-Environment Interaction with EMG Signals: A Review, in IEEE/CAA journal of automatica sinica, DOI: 10.1109/JAS.2024.124329.
Keywords: Algorithms; Control systems; Electromyography; Heuristic algorithms; Human engineering; Human-robot interaction; human-robot interaction (HRI); human-robot-environment interaction (HREI); Myoelectricity; Process control; Reinforcement learning; Reviews; Robot sensing systems; Robots; semiautonomous; Sensory feedback.
Abstract: A long history has passed since electromyography (EMG) signals have been explored in human-centered robots for intuitive interaction. However, it still has a gap between scientific research and real-life applications. Previous studies mainly focused on EMG decoding algorithms, leaving a dynamic relationship between the human, robot, and uncertain environment in real-life scenarios seldomly concerned. To fill this gap, this paper presents a comprehensive review of EMG-based techniques in human-robot-environment interaction (HREI) systems. The general processing framework is summarized, and three interaction paradigms, including direct control, sensory feedback, and partial autonomous control, are introduced. EMG-based intention decoding is treated as a module of the proposed paradigms. Five key issues involving precision, stability, user attention, compliance, and environmental awareness in this field are discussed. Several important directions, including EMG decomposition, robust algorithms, HREI dataset, proprioception feedback, reinforcement learning, and embodied intelligence, are proposed to pave the way for future research. To the best of what we know, this is the first time that a review of EMG-based methods in the HREI system is summarized. It provides a novel and broader perspective to improve the practicability of current myoelectric interaction systems, in which factors in human-robot interaction, robot-environment interaction, and state perception by human sensations are considered, which has never been done by previous studies.

(2024) Passive Model Predictive Impedance Control for Safe Physical Human-Robot Interaction, in IEEE Trans. Cogn. Dev. Syst., DOI: 10.1109/TCDS.2023.3275217.
Keywords: Aerospace electronics; Closed loops; Control methods; Control systems design; Controllers; Human behavior; Human motion; Human-robot interaction; Impedance; impedance control; Internal energy; model predictive control; passivity; Physical human-robot interaction; Predictive control; Predictive models; Robots; Stiffness; Task analysis; Torque.
Abstract: Various cognitive systems have been designed to model the position and stiffness profiles of human behavior and then to drive robots by mimicking the human’s behavior to accomplish physical human-robot interaction tasks through a properly designed impedance controller. However, some studies have shown that variable stiffness parameters of the impedance controller can cause the violation of the passivity constraint of the robot states, and make the robot’s stored energy exceed the external energy injected from the human user, thus leading to the unsafe human-robot interaction. To solve this problem, this paper proposes a novel passive model predictive impedance control method including two control loops. In the bottom-loop of the proposed controller, the robot is driven by a variable impedance controller to achieve the desired compliant interaction behavior. In the top-loop of the proposed controller, the model predictive control (MPC) is used to ensure that the robot states satisfy the passivity constraint by calculating a complementary torque to limit the stored energy of the robot. The passivity of the closed-loop robot system and the feasibility of MPC are guaranteed by theoretical analysis, ensuring the safety of the robotic movement in the human-robot interaction. The effectiveness of the proposed method is demonstrated by the simulation and experiment on the Franka Emika Panda robot.;Various cognitive systems have been designed to model the position and stiffness profiles of human behavior and then to drive robots by mimicking the human’s behavior to accomplish physical human–robot interaction tasks through a properly designed impedance controller. However, some studies have shown that variable stiffness parameters of the impedance controller can cause the violation of the passivity constraint of the robot states, and make the robot’s stored energy exceed the external energy injected from the human user, thus leading to the unsafe human–robot interaction. To solve this problem, this article proposes a novel passive model-predictive impedance control method including two control loops. In the bottom-loop of the proposed controller, the robot is driven by a variable impedance controller to achieve the desired compliant interaction behavior. In the top-loop of the proposed controller, the model-predictive control (MPC) is used to ensure that the robot states satisfy the passivity constraint by calculating a complementary torque to limit the stored energy of the robot. The passivity of the closed-loop robot system and the feasibility of MPC are guaranteed by theoretical analysis, ensuring the safety of the robotic movement in the human–robot interaction. The effectiveness of the proposed method is demonstrated by the simulation and experiment on the Franka Emika Panda robot.

(2024) The Human-Machine Interaction Methods and Strategies for Upper and Lower Extremity Rehabilitation Robots: A Review, in IEEE Sens. J., DOI: 10.1109/JSEN.2024.3374344.
Keywords: Assistive robots; Control methods; Control strategies; Electroencephalography; Electromyography; Human computer interaction; human intention recognition; human-machine interaction (HMI); Interactive control; Recognition; Rehabilitation; rehabilitation robot; Rehabilitation robots; Robot control; Robot sensing systems; Robotics; Robots; sensing methods; Sensors; Training.
Abstract: The development of intelligent rehabilitation robots has greatly reduced the workload of rehabilitation physicians. Human-machine interaction (HMI) control methods are a critical technology for intelligent rehabilitation robots. Therefore, we systematically review the HMI methods and control strategies for upper and lower limb rehabilitation robots and summarizing the HMI methods with different sensors. The integration of rehabilitation robots and HMI control methods has grown significantly in recent years. For this reason, this article takes the sensing methods as the entry point to give readers a quick overview of the current status of HMI research. We present different sensing methods, interactive control strategies, applications, and evaluation methods and discuss the limitations and future development directions in the field. The results show that the mainstream control methods of HMI are based on motion signals, surface electromyography (sEMG), ultrasound (US), and electroencephalogram (EEG). In the field of rehabilitation robotics, human intention recognition-based interaction strategy is the mainstream HMI strategy, which mainly collects biosignals, force/moment, spatial angle, and other information for human intention recognition. Future research may focus on the use of multimodal sensing interactions, flexible control strategies, and generalized rehabilitation assessment mechanism.

(2024) Human-Like Trajectory Planning Based on Postural Synergistic Kernelized Movement Primitives for Robot-Assisted Rehabilitation, in IEEE T. Hum.-Mach. Syst., DOI: 10.1109/THMS.2024.3360111.
Keywords: Assistive robots; Coordination; Human motion; Kernelized movement primitives; Motion capture; Motors; postural synergy; Posture; Principal component analysis; Principal components analysis; Rehabilitation; rehabilitation robot; Rehabilitation robots; Robot kinematics; Robots; Subspaces; Trajectory planning.
Abstract: The motor synergy pattern is an intrinsic characteristic found in natural human movements, particularly in the upper limb. It is essential to improve the multijoint coordination ability for stroke patients by integrating the synergy pattern into rehabilitation tasks and trajectory design. However, current robot-assisted rehabilitation systems tend to overlook the incorporation of a multijoint synergy model. This article proposes postural synergistic kernelized movement primitives (PSKMP) method for the human-like trajectory planning of robot-assisted upper limb rehabilitation. First, the demonstrated trajectory obtained from the motion capture system is subject to principal component analysis to extract postural synergies. Then, the PSKMP is proposed by kernelizing the postural synergistic subspaces with the kernel treatment to preserve human natural movement characteristics. Finally, the rehabilitation trajectory accord with human motion habits can be generated based on generalized postural synergistic subspaces. This approach has undergone practical validation on an upper limb rehabilitation robot, and the experimental results show that the proposed method enables the generation of human-like trajectories adapted to new task points, in accordance with the natural movement style of human. This method holds great significance in promoting the recovery of coordination ability of stroke patients.

(2024) Uncertainty Compensated High-Order Adaptive Iteration Learning Control for Robot-Assisted Upper Limb Rehabilitation, in IEEE Trans. Autom. Sci. Eng., DOI: 10.1109/TASE.2023.3335401.
Keywords: Adaptation models; Assistive robots; Iterative learning control; Limbs; model-free control; Nonlinear dynamical systems; Robots; Training; Trajectory; Uncertainty; uncertainty compensation; Upper limb rehabilitation.
Abstract: Upper limb rehabilitation robot can assist stroke patients to complete daily activities to promote the recovery of upper-limb motor functions. However, the robot uncertainty and the patient’s unconscious disturbance impose great difficulties on the high-performance trajectory tracking of the rehabilitation robot. In this paper, an uncertainty compensated high-order adaptive iterative learning controller (UCHAILC) is proposed to reduce the impact of uncertainty from inside and outside of the robot during the rehabilitation process. The nonlinear system is converted into a dynamic linearization model with uncertainty compensation, and the optimization criterion method is adopted to estimate the pseudo-partial derivative (PPD) parameters and the uncertainty respectively, then the previous iterations are used to update the current parameters through a high-order learning scheme. The convergence of UCHAILC is theoretically proved. Simulation and control experiments on a rehabilitation robot are given to validate the effectiveness of the proposed method, which is significant to improve the training security and physiotherapy effect of robot-assisted rehabilitation. Note to Practitioners-This paper was motivated by the need to assist stroke patients to restore motor function for executing daily activities. The inherent difficulties lie in reducing the tracking errors of rehabilitation robots caused by uncertainty and involuntary disturbance from patients to avoid secondary injury. The proposed UCHAILC can transform the complex nonlinear system into a dynamic linear model with uncertainty compensation, then the PPD parameters and uncertainty are estimated through high-order learning law. Theoretical analysis, simulation, and experiments verified the feasibility of the method. Furthermore, the proposed controller is not limited to the dynamic model and hardware driving mode of the robot system, which can be easily transplanted to other nonlinear control systems with uncertainties.

(2024) Differential Game-Based Control for Nonlinear Human-Robot Interaction System With Unknown Desired Trajectory, in IEEE T. Cybern., DOI: 10.1109/TCYB.2024.3402353.
Keywords: Algorithms; Computer Simulation; Differential flatness; differential game; Differential games; Game Theory; Human-robot interaction; human-robot interaction (HRI); Humans; Man-Machine Systems; Nonlinear Dynamics; nonlinear system; Nonlinear systems; probabilistic error bound; Probabilistic logic; Robotics - methods; Robots; Trajectory; Trajectory tracking.
Abstract: Differential game is an effective technique to describe the negotiation between the humans and robots, which is widely used to realize the trajectory tracking tasks in the human-robot interaction (HRI). However, most existing works consider the control-affine HRI systems and assume the desired trajectory is available to both the human and the robot, which limit the scope of applications. To overcome these difficulties, this work focuses on the nonaffine HRI system and supposes that the desired trajectory is not available to the robot. A novel differential game framework encoding the desired trajectory estimator is proposed, where the desired trajectory is estimated via the Gaussian process regression (GPR) technique. To address the challenge arising from the nonlinearity of the HRI system, we equivalently transform the original problem into the one in a differentially flat space, and seek the equilibrium strategies for the transformed problem substitutionally. We further prove that the trajectory tracking error satisfies a probabilistic bound, whose confidence interval tightens as the decrease of noise variance during the interaction. Comparative simulation results show that our method outperforms the learning-based method in terms of robustness, parameters setting, and time consumption. Experiment results further show that the tracking error under the proposed human-robot cooperative algorithm is reduced by 55% compared to the human direct control.Differential game is an effective technique to describe the negotiation between the humans and robots, which is widely used to realize the trajectory tracking tasks in the human-robot interaction (HRI). However, most existing works consider the control-affine HRI systems and assume the desired trajectory is available to both the human and the robot, which limit the scope of applications. To overcome these difficulties, this work focuses on the nonaffine HRI system and supposes that the desired trajectory is not available to the robot. A novel differential game framework encoding the desired trajectory estimator is proposed, where the desired trajectory is estimated via the Gaussian process regression (GPR) technique. To address the challenge arising from the nonlinearity of the HRI system, we equivalently transform the original problem into the one in a differentially flat space, and seek the equilibrium strategies for the transformed problem substitutionally. We further prove that the trajectory tracking error satisfies a probabilistic bound, whose confidence interval tightens as the decrease of noise variance during the interaction. Comparative simulation results show that our method outperforms the learning-based method in terms of robustness, parameters setting, and time consumption. Experiment results further show that the tracking error under the proposed human-robot cooperative algorithm is reduced by 55% compared to the human direct control.;Differential game is an effective technique to describe the negotiation between the humans and robots, which is widely used to realize the trajectory tracking tasks in the human-robot interaction (HRI). However, most existing works consider the control-affine HRI systems and assume the desired trajectory is available to both the human and the robot, which limit the scope of applications. To overcome these difficulties, this work focuses on the nonaffine HRI system and supposes that the desired trajectory is not available to the robot. A novel differential game framework encoding the desired trajectory estimator is proposed, where the desired trajectory is estimated via the Gaussian process regression (GPR) technique. To address the challenge arising from the nonlinearity of the HRI system, we equivalently transform the original problem into the one in a differentially flat space, and seek the equilibrium strategies for the transformed problem substitutionally. We further prove that the trajectory tracking error satisfies a probabilistic bound, whose confidence interval tightens as the decrease of noise variance during the interaction. Comparative simulation results show that our method outperforms the learning-based method in terms of robustness, parameters setting, and time consumption. Experiment results further show that the tracking error under the proposed human-robot cooperative algorithm is reduced by 55% compared to the human direct control.

(2024) Virtually Constrained Admittance Control Using Feedback Linearization for Physical Human-Robot Interaction With Rehabilitation Exoskeletons, in IEEE/ASME transactions on mechatronics, DOI: 10.1109/TMECH.2024.3480157.
Keywords: Admittance control; Aerospace electronics; Exoskeletons; Feedback linearization; Force; holonomic constraints; Mechatronics; physical human–robot interaction (pHRI); rehabilitation exoskeletons; Robot kinematics; Robots; Sun; Trajectory.
Abstract: Robot-assisted rehabilitation focuses in part on path-based assist-as-needed reaching rehabilitation, which dynamically adapts the level of robot assistance during physical therapy to ensure patient progress along a predefined trajectory without overreliance on the system. Additionally, bimanual exoskeletons have enabled asymmetric rehabilitation schemes, which leverage the patient’s healthy side to guide the rehabilitation through interactions with objects in virtual reality that replicate activities of daily living. Within the context of physical human-robot interaction, these tasks can be formulated as constraints on the space of allowable motions. This study introduces a novel feedback linearization-inspired time-invariant admittance control scheme that enforces these motion constraints by isolating and stabilizing the component of the virtual dynamics transversal to the constraint. The methodology is applied to two rehabilitation tasks: 1) a path-guided reaching task with restoring force field and 2) a bimanual interaction with a virtual object. Each task is then evaluated on one of two drastically different exoskeleton systems: 1) the V-Rex, a nonanthropomorphic full-body haptic device and 2) the EXO-UL8, an anthropomorphic bimanual upper-limb exoskeleton. The two systems exist on opposite ends of the task/joint space control, nonredundant/redundant, off-the-shelf (industrial)/custom, nonanthropomorphic/anthropomorphic spectra. Experimental results validate and support the methodology as a generalizable approach to enabling constrained admittance control for rehabilitation robots.

(2024) Simultaneous Estimation of Human Motion Intention and Time-Varying Arm Stiffness for Enhanced Human-Robot Interaction, in IEEE Trans. Cogn. Dev. Syst., DOI: 10.1109/TCDS.2024.3480854.
Keywords: End effectors; Estimation; Force; Human arm stiffness; Human-robot interaction; Impedance; Motion intention; Motor drives; Physical human-robot interaction; Robot kinematics; Robot sensing systems; Robots; Variable impedance control; Vectors.
Abstract: Recent advances in physiological human motor control research indicate that human endpoint stiffness magnitude increases linearly with grasp force. Based on these findings, a scheme was proposed in this paper to integrate the linear quadratic estimation (LQE) filter with the stiffness model inferred from grasp force, which can simultaneously estimate the human arm’s stiffness and motion intention. Then, an online variable impedance controller (VIC) was designed based on previous estimations for physical human-robot interaction (pHRI). The proposed stiffness model and estimation method were validated through experiments using a planar robotic interface. In order to assess its performance in practical pHRI tasks, the implementation of human arm stiffness and intention estimation combining with VIC was extended to a teleoperation peg-in-hole and robot-assisted rehabilitation tasks. The experimental results demonstrate that the proposed method can effectively estimate human motion intention and arm stiffness simultaneously. Compared to existing methods, the proposed VIC enhances pHRI in terms of increased flexibility, effective guidance, and reduced human effort.

(2024) Design of Human–Machine Compatible Ankle Rehabilitation Robot Based on Equivalent Human Ankle Model, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2023.3342556.
Keywords: Analytical models; Ankle; ankle rehabilitation robot; Assistive robots; center-of-rotation estimation; Degrees of freedom; Equivalence; equivalent model; Estimation; Human motion; Human-machine systems; Human–machine compatibility; Kinematics; mechanical design principles; Mechanical systems; Medical robotics; Principles; Real time; Real-time systems; Rehabilitation; Rehabilitation robots; Robots; Rotation; Sockets; Workspace.
Abstract: In this letter, a human–machine compatible ankle rehabilitation robot (HMCARR) is proposed to help stroke patients with motion dysfunction recover their motor function. The HMCARR can make the human ankle center-of-rotation (H-CoR) and the ankle rehabilitation robot center-of-rotation (R-CoR) coincide in real-time. The overall idea is to analyze from the “human” perspective to the “human–machine” perspective, and then to the “machine” perspective. Firstly, from the “human” perspective: a spherical-pair-with-clearance equivalent human ankle model and a H-CoR estimation model are proposed, and the motion range of the H-CoR is analyzed. Secondly, from the “ human–machine” perspective: to obtain the mechanical design principles of the HMCARR, degree-of-freedom (DOF) and kinematic analyses of the human–machine closed chain model are conducted. Finally, from the “machine” perspective: DOF, kinematics, singularity, sensitivity, and workspace analysis are performed for the HMCARR based on a 3-RRCRR parallel mechanism, where R and C represent a revolute pair and a cylindrical pair, respectively. This study indicates that the mechanical design principles of the HMCARR include three requirements for the DOFs, kinematic independence, and position workspace. The results suggest that the HMCARR based on the 3-RRCRR parallel mechanism can achieve coincidence between H-CoR and R-CoR in real-time.

(2024) A Smooth Velocity Transition Framework Based on Hierarchical Proximity Sensing for Safe Human-Robot Interaction, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2024.3385608.
Keywords: Acceleration; Capacitive sensors; Collision avoidance; Fifth Industrial Revolution; Human engineering; Human-robot interaction; Industrial safety; Industry 5.0; Proximity; Robot arms; Robot sensing systems; robot skin; Robots; Safety; smooth velocity transition framework; the safety of humans; Velocity; Velocity control.
Abstract: With the rapid technology development pushing the introduction of the fifth industrial revolution, Industry 5.0, robots are getting rid of fences and sharing the workspace with humans. In such a context, ensuring the safety of humans and robots is a critical demand. One of the effective methods for this is proximity sensing, among which capacitive sensors are widely used to detect the proximity of humans. However, the capacitive sensor cannot get accurate distance information since the capacitance varies with the characteristics of obstacles. This work develops a capacitive robot skin, seamlessly integrated into the proposed smooth velocity transition framework to deal with this challenge. The robot skin is customized to cover a large area on the exterior of a 6-DOF robot arm. A hierarchical proximity perception approach is used to grade the sensing state. Based on this, distance reduction and collision avoidance velocity generation methods are used to reach a smooth and quick decay of the velocity. The control strategy is applied in a pick-and-place scenery for verification. Compared to the traditional threshold trigger method, the proposed smooth velocity transition framework can greatly reduce the absolute value of the local maximum acceleration, which can enable a flexible and natural human-robot interaction while ensuring human safety.

(2024) Haptic Transparency and Interaction Force Control for a Lower-Limb Exoskeleton, in IEEE Trans. Robot., DOI: 10.1109/TRO.2024.3359541.
Keywords: assistive robots; Closed loops; Constraints; Controllers; Dynamic models; Dynamics; Exoskeletons; Foot; Force measurement; Gait; interaction force control; Legged locomotion; Physical human-robot interaction; rehabilitation robots; Robots; Sensors; Torque; Torquemeters; Tracking control.
Abstract: Controlling the interaction forces between a human and an exoskeleton is crucial for providing transparency or adjusting assistance or resistance levels. However, it is an open problem to control the interaction forces of lower limb exoskeletons designed for unrestricted overground walking. For these types of exoskeletons, it is challenging to implement force/torque sensors at every contact between the user and the exoskeleton for direct force measurement. Moreover, it is important to compensate for the exoskeleton’s whole-body gravitational and dynamical forces, especially for heavy lower limb exoskeletons. Previous works either simplified the dynamic model by treating the legs as independent double pendulums, or they did not close the loop with interaction force feedback. The proposed whole-exoskeleton closed-loop compensation (WECC) method calculates the interaction torques during the complete gait cycle by using whole-body dynamics and joint torque measurements on a hip-knee exoskeleton. Furthermore, it uses a constrained optimization scheme to track desired interaction torques in a closed loop while considering physical and safety constraints. We evaluated the haptic transparency and dynamic interaction torque tracking of WECC control on three subjects. We also compared the performance of WECC with a controller based on a simplified dynamic model and a passive version of the exoskeleton. The WECC controller results in a consistently low absolute interaction torque error during the whole gait cycle for both zero and nonzero desired interaction torques. In contrast, the simplified controller yields poor performance in tracking desired interaction torques during the stance phase.;Controlling the interaction forces between a human and an exoskeleton is crucial for providing transparency or adjusting assistance or resistance levels. However, it is an open problem to control the interaction forces of lower-limb exoskeletons designed for unrestricted overground walking. For these types of exoskeletons, it is challenging to implement force/torque sensors at every contact between the user and the exoskeleton for direct force measurement. Moreover, it is important to compensate for the exoskeleton’s whole-body gravitational and dynamical forces, especially for heavy lower-limb exoskeletons. Previous works either simplified the dynamic model by treating the legs as independent double pendulums, or they did not close the loop with interaction force feedback. The proposed whole-exoskeleton closed-loop compensation (WECC) method calculates the interaction torques during the complete gait cycle by using whole-body dynamics and joint torque measurements on a hip-knee exoskeleton. Furthermore, it uses a constrained optimization scheme to track desired interaction torques in a closed loop while considering physical and safety constraints. We evaluated the haptic transparency and dynamic interaction torque tracking of WECC control on three subjects. We also compared the performance of WECC with a controller based on a simplified dynamic model and a passive version of the exoskeleton. The WECC controller results in a consistently low absolute interaction torque error during the whole gait cycle for both zero and nonzero desired interaction torques. In contrast, the simplified controller yields poor performance in tracking desired interaction torques during the stance phase.

(2024) Toward Task-Independent Optimal Adaptive Control of a Hip Exoskeleton for Locomotion Assistance in Neurorehabilitation, in IEEE transactions on systems, man, and cybernetics. Systems, DOI: 10.1109/TSMC.2024.3454556.
Keywords: Adaptive control; Assistance personalization; Assistive robots; Exoskeletons; Impedance; impedance control; Legged locomotion; Neurorehabilitation; optimal adaptive control; Patient rehabilitation; rehabilitation exoskeletons; rehabilitation robotics; Reinforcement learning; reinforcement learning (RL); Tuning.
Abstract: Personalized robotic exoskeleton control is essential in assisting individuals with motor deficits. However, current research still lacks a solution from the end of a practical need of the problem to the end of its successful demonstration in physical environments, namely an end-to-end solution, that enables stable and continuous walking across different tasks. This study addresses this challenge by introducing a hierarchical control framework for the purpose. At the low level, impedance control ensures joint compliance without causing injury to users. At the high level, a reinforcement learning (RL)-based optimal adaptive controller automatically personalizes assistance to both hip extension and flexion (namely, bi-directional) to reach a target range of motion (ROM) under multiple walking conditions. As the first potentially feasible approach to this challenging problem and to meet practical use requirements, we developed a least-square policy iteration-based solution to configure the intrinsic parameters within the well-established finite state machine impedance control (FSM-IC). We successfully tested the control solution on eight young unimpaired participants and one participant post-stroke wearing a hip exoskeleton while walking on an instrumented treadmill. The proposed method can be applied to solving for optimal impedance parameters for individual users and different task scenarios to increase joint ROM. Our next step is to further evaluate this solution framework on additional people with hemiparesis who may benefit from hip joint assistance in therapy or daily activities to restore normative or improve gait patterns.

(2024) Task Space Compliant Control and Six-Dimensional Force Regulation Toward Automated Robotic Ultrasound Imaging, in IEEE Trans. Autom. Sci. Eng., DOI: 10.1109/TASE.2023.3282974.
Keywords: admittance control; Aerospace electronics; coupled stability; Force; force control; Imaging; Probes; Robot sensing systems; Robotic ultrasound; Robots; Task analysis.
Abstract: Objective: We propose a general control framework for task space compliant motion and six-dimensional (6-D) force regulation towards automated robotic ultrasound (US) imaging. The framework endows a position-controlled robotic manipulator with the capability of accurate compliant motion in free space and accurate force control in motion-constrained environment. Methods: An intuitive six degree-of-freedom (6-DoF) admittance control model expressed in an arbitrary Cartesian body frame is mathematically derived with closed-form task space error mapping. Its practical implementation on widely-used collaborative manipulators is proposed to achieve full task space compliant behaviors and accurate 6-D force control. A hybrid control law is presented to achieve good motion accuracy in free space and improved coupled stability in motion-constrained environment. The coupled model of physical human-robot interaction is established and the reason for the improved coupled stability is analyzed through simulation. Results: Evaluation experiments on the proposed control framework were performed to show the effectiveness. The mean error of compliant trajectory following was less than 0.30 mm in free space. The mean relative force and moment control accuracy in three orthogonal directions was better than 0.5% and 0.8%, respectively. The improved coupled stability under the same model parameters was also confirmed by human-robot interaction experiments. Finally, an automated robotic US imaging experiment on a human volunteer in a real clinical scenario was carried out to show the potential application of our proposed framework. Conclusion: Experimental results have shown the advantages of the control framework, including satisfied force control accuracy, high accuracy of compliant motion, improved coupled stability, and system effectiveness on a human volunteer. Note to Practitioners-This paper was motivated by the increasing needs of automated ultrasound (US) scanning for both diagnostic and interventional purpose. Clinical sonographers suffer from repeated workload when performing diagnostic US imaging, which could benefit from automated robotic scanning. Robotic US imaging involves physical interaction between the robot end-effector (i.e., US transducer) and the human body. The dynamics of the interaction is regulated by the control law to guarantee the contact of the US transducer and the safety of the procedure. Most existing works have focused on regulating in-plane contact force in terms of the position without considering the compliance in other dimensions. However, it is not a trivial work to extend the positional compliance to six degree-of-freedom (6-DoF) compliance. As the prevalence of low-cost collaborative robotic arms in medical scenarios, how to perform 6-DoF compliant trajectory following and accurate six-dimensional (6-D) force control on these robotic arms becomes increasingly important. This paper gives a complete general solution to achieve 6-DoF compliant control and 6-D force regulation with accurate kinematics on a position-controlled robotic arm. A hybrid control law is proposed to switch the government of “instantaneous model” and “theoretical model” to achieve compliant motion accuracy in free space and improved coupled stability in motion-constrained environment. No expensive torque sensors and torque control interface are required. And no prior geometric knowledge about the scanning object is needed. We have demonstrated the application for robotic US imaging in a real clinical scenario.

(2024) Design and Quantitative Assessment of Teleoperation-Based Human–Robot Collaboration Method for Robot-Assisted Sonography, in IEEE Trans. Autom. Sci. Eng., DOI: 10.1109/TASE.2024.3350524.
Keywords: Robots; Ultrasonic imaging; Task analysis; Collaboration; Haptic interfaces; Image quality; Torque; Robot-assisted sonography; continuous and intuitive teleoperation; teleoperation control; human–robot collaboration; human–robot interface.
Abstract: Tele-echography has emerged as a promising and effective solution, leveraging the expertise of sonographers and the autonomy of robots to perform ultrasound scanning for patients residing in remote areas, without the need for in-person visits by the sonographer. Designing effective and natural human-robot interfaces for tele-echography remains challenging, with patient safety being a critical concern. In this article, we develop a teleoperation system for robot-assisted sonography with two different interfaces, a haptic device-based interface and a low-cost 3D Mouse-based interface, which can achieve continuous and intuitive telemanipulation by a leader device with a small workspace. To achieve compliant interaction with patients, we design impedance controllers in Cartesian space to track the desired position and orientation for these two teleoperation interfaces. We also propose comprehensive evaluation metrics of robot-assisted sonography, including subjective and objective evaluation, to evaluate tele-echography interfaces and control performance. We evaluate the ergonomic performance based on the estimated muscle fatigue and the acquired ultrasound image quality. We conduct user studies based on the NASA Task Load Index to evaluate the performance of these two human-robot interfaces. The tracking performance and the quantitative comparison of these two teleoperation interfaces are conducted by the Franka Emika Panda robot. The results and findings provide guidance on human-robot collaboration design and implementation for robot-assisted sonography. Note to Practitioners—Robot-assisted sonography has demonstrated efficacy in medical diagnosis during clinical trials. However, deploying fully autonomous robots for ultrasound scanning remains challenging due to various constraints in practice, such as patient safety, dynamic tasks, and environmental uncertainties. Semi-autonomous or teleoperation-based robot sonography represents a promising approach for practical deployment. Previous work has produced various expensive teleoperation interfaces but lacks user studies to guide teleoperation interface selection. In this article, we present two typical teleoperation interfaces and implement a continuous and intuitive teleoperation control system. We also propose a comprehensive evaluation metric for assessing their performance. Our findings show that the haptic device outperforms the 3D Mouse, based on operators’ feedback and acquired image quality. However, the haptic device requires more learning time and effort in the training stage. Furthermore, the developed teleoperation system offers a solution for shared control and human-robot skill transfer. Our results provide valuable guidance for designing and implementing human-robot interfaces for robot-assisted sonography in practice.

(2024) Force Sensing and Compliance Control for a Cable-Driven Redundant Manipulator, in IEEE/ASME transactions on mechatronics, DOI: 10.1109/TMECH.2023.3263922.
Keywords: Admittance control; Algorithms; cable-driven redundant manipulator (CDRM); Cables; Compliance; Confined spaces; Controllers; dynamic model; Dynamics; End effectors; Force; force sensing; Friction; human–robot interaction; Kinematics; Manipulator dynamics; Manipulators; Modular design; Robot sensing systems; Robots; Sensors; Torque sensors (robotics).
Abstract: The cable-driven redundant manipulator (CDRM) has a light slender and highly dexterous body, which is especially suitable for fine manipulation in confined spaces or unstructured environments. However, for dexterous manipulation and physical interaction with its surrounding environment, the force/torque sensor is indispensable, which will significantly increase the mass, dimension, and cost of the whole CDRM system. In this article, a force-sensing algorithm and compliance control framework for CDRM without a six-axis force/torque sensor are proposed. First, we design a modular cable-driven manipulator with two-level internal sensors, i.e., joint encoders and cable tension sensors. The multispace kinetic model is derived to establish the mapping between motor, cable, joint, and end-effector states. At the same time, we build a recursive dynamics model that takes the cables’ tension, cable–hole friction, links’ gravity, and end-effector forces into account. Then, the indirect force-sensing algorithm of the end-effector is proposed by combining the kinematic and dynamic equations and internal sensor information. Furthermore, a compliance controller based on indirect force sensing is designed. Finally, typical experiments are carried out based on the CDRM. Experiment results indicate that the force-sensing accuracy exceeds 95%, whereas the compliance controller demonstrates outstanding compliant behavior in human–robot interaction tasks.

(2024) Effectiveness of Intelligent Control Strategies in Robot-Assisted Rehabilitation-A Systematic Review, in IEEE Trans. Neural Syst. Rehabil. Eng., DOI: 10.1109/TNSRE.2024.3396065.
Keywords: Adaptability; Algorithms; Artificial Intelligence; Assistive robots; Control systems; Control systems design; Controllers; Criteria; human joints; Humans; Intelligent control; machine intelligence; Machine learning; Optimization; physical human-robot interaction; Rehabilitation; Rehabilitation - instrumentation; Rehabilitation - methods; Rehabilitation robots; Reviews; Robot control; Robot learning; Robot sensing systems; Robotics; Robots; System effectiveness; Treatment Outcome.
Abstract: This review aims to provide a systematic analysis of the literature focused on the use of intelligent control systems in robotics for physical rehabilitation, identifying trends in recent research and comparing the effectiveness of intelligence used in control, with the aim of determining important factors in robot-assisted rehabilitation and how intelligent controller design can improve them. Seven electronic research databases were searched for articles published in the years 2015 - 2022 with articles selected based on relevance to the subject area of intelligent control systems in rehabilitation robotics. It was found that the most common use of intelligent algorithms for control is improving traditional control strategies with optimization and learning techniques. Intelligent algorithms are also commonly used in sensor output mapping, model construction, and for various data learning purposes. Experimental results show that intelligent controllers consistently outperform non-intelligent controllers in terms of transparency, tracking accuracy, and adaptability. Active participation of the patients and lowered interaction forces are consistently mentioned as important factors in improving the rehabilitation outcome as well as the patient experience. However, there are limited examples of studies presenting experimental results with impaired participants suffering limited range of motion, so the effectiveness of therapy provided by these systems is often difficult to quantify. A lack of universal evaluation criteria also makes it difficult to compare control systems outside of articles which use their own comparison criteria.This review aims to provide a systematic analysis of the literature focused on the use of intelligent control systems in robotics for physical rehabilitation, identifying trends in recent research and comparing the effectiveness of intelligence used in control, with the aim of determining important factors in robot-assisted rehabilitation and how intelligent controller design can improve them. Seven electronic research databases were searched for articles published in the years 2015 - 2022 with articles selected based on relevance to the subject area of intelligent control systems in rehabilitation robotics. It was found that the most common use of intelligent algorithms for control is improving traditional control strategies with optimization and learning techniques. Intelligent algorithms are also commonly used in sensor output mapping, model construction, and for various data learning purposes. Experimental results show that intelligent controllers consistently outperform non-intelligent controllers in terms of transparency, tracking accuracy, and adaptability. Active participation of the patients and lowered interaction forces are consistently mentioned as important factors in improving the rehabilitation outcome as well as the patient experience. However, there are limited examples of studies presenting experimental results with impaired participants suffering limited range of motion, so the effectiveness of therapy provided by these systems is often difficult to quantify. A lack of universal evaluation criteria also makes it difficult to compare control systems outside of articles which use their own comparison criteria.;This review aims to provide a systematic analysis of the literature focused on the use of intelligent control systems in robotics for physical rehabilitation, identifying trends in recent research and comparing the effectiveness of intelligence used in control, with the aim of determining important factors in robot-assisted rehabilitation and how intelligent controller design can improve them. Seven electronic research databases were searched for articles published in the years 2015 - 2022 with articles selected based on relevance to the subject area of intelligent control systems in rehabilitation robotics. It was found that the most common use of intelligent algorithms for control is improving traditional control strategies with optimization and learning techniques. Intelligent algorithms are also commonly used in sensor output mapping, model construction, and for various data learning purposes. Experimental results show that intelligent controllers consistently outperform non-intelligent controllers in terms of transparency, tracking accuracy, and adaptability. Active participation of the patients and lowered interaction forces are consistently mentioned as important factors in improving the rehabilitation outcome as well as the patient experience. However, there are limited examples of studies presenting experimental results with impaired participants suffering limited range of motion, so the effectiveness of therapy provided by these systems is often difficult to quantify. A lack of universal evaluation criteria also makes it difficult to compare control systems outside of articles which use their own comparison criteria.

(2024) Trajectory Deformation-Based Multi-Modal Adaptive Compliance Control for a Wearable Lower Limb Rehabilitation Robot, in IEEE Trans. Neural Syst. Rehabil. Eng., DOI: 10.1109/TNSRE.2023.3348332.
Keywords: Trajectory; Legged locomotion; Torque; Training; Assistive robots; Deformation; Control systems; Wearable lower limb rehabilitation robot; physical human-robot interaction; trajectory deformation algorithm; dynamic motion primitives; linear active disturbance rejection control.
Abstract: Adaptive compliance control is critical for rehabilitation robots to cope with the varying rehabilitation needs and enhance training safety. This article presents a trajectory deformation-based multi-modal adaptive compliance control strategy (TD-MACCS) for a wearable lower limb rehabilitation robot (WLLRR), which includes a high-level trajectory planner and a low-level position controller. Dynamic motion primitives (DMPs) and a trajectory deformation algorithm (TDA) are integrated into the high-level trajectory planner, generating multi-joint synchronized desired trajectories through physical human-robot interaction (pHRI). In particular, the amplitude modulation factor of DMPs and the deformation factor of TDA are adapted by a multi-modal adaptive regulator, achieving smooth switching of human-dominant mode, robot-dominant mode, and soft-stop mode. Besides, a linear active disturbance rejection controller is designed as the low-level position controller. Four healthy participants and two stroke survivors are recruited to conduct robot-assisted walking experiments using the TD-MACCS. The results show that the TD-MACCS can smoothly switch three control modes while guaranteeing trajectory tracking accuracy. Moreover, we find that appropriately increasing the upper bound of the deformation factor can enhance the average walking speed (AWS) and root mean square of trajectory deviation (RMSTD).

(2024) Dynamic Motion Primitives-Based Trajectory Learning for Physical Human-Robot Interaction Force Control, in IEEE Trans. Ind. Inform., DOI: 10.1109/TII.2023.3280320.
Keywords: Behavioral sciences; Control methods; Dynamic motion primitives (DMPs); Force; Force control; Human-robot interaction; interaction force; Interactive control; iterative learning (IL); Iterative methods; Learning; Performance degradation; physical human–robot interaction (pHRI); Rehabilitation robots; Robot control; Robot dynamics; Robot kinematics; Robots; Trajectories; Trajectory.
Abstract: One promising function of interactive robots is to provide a specific interaction force to human users. For example, rehabilitation robots are expected to promote patients’ recovery by interacting with them with a prescribed force. However, motion uncertainties of different individuals, which are hard to predict due to the varying motion speed and noises during motion, degrade the performance of existing control methods. This article proposes a method to learn a desired reference trajectory for a robot based on dynamic motion primitives (DMPs) and iterative learning (IL). By controlling the robot to follow the generated desired reference trajectory, the interaction force can achieve a desired value. In our proposed approach, DMPs are first employed to parameterize the demonstration trajectories of the human user. Then, a recursive least square (RLS)-based estimator is developed and combined with the Adam optimization method to update the trajectory parameters so that the desired reference trajectory of the robot is iteratively obtained by resolving the DMPs. Since the proposed method parameterizes the trajectories depending on the phase variable, it removes the essential assumption of traditional IL methods that the iteration period should be invariant, and thus, has improved robustness compared with the existing methods. Experiments are performed using an interactive robot to validate the effectiveness of our proposed scheme.

(2024) Active Neural Network Control for a Wearable Upper Limb Rehabilitation Exoskeleton Robot Driven by Pneumatic Artificial Muscles, in IEEE Trans. Neural Syst. Rehabil. Eng., DOI: 10.1109/TNSRE.2024.3429206.
Keywords: Hysteresis; Artificial neural networks; Robots; Exoskeletons; Approximation error; Accuracy; Training; Upper limb rehabilitation; pneumatic artificial muscle; unscented Kalman filter; multilayer feedforward neural network; hysteresis compensation.
Abstract: Pneumatic artificial muscle (PAM) has been widely used in rehabilitation and other fields as a flexible and safe actuator. In this paper, a PAM-actuated wearable exoskeleton robot is developed for upper limb rehabilitation. However, accurate modeling and control of the PAM are difficult due to complex hysteresis. To solve this problem, this paper proposes an active neural network method for hysteresis compensation, where a neural network (NN) is utilized as the hysteresis compensator and unscented Kalman filtering is used to estimate the weights and approximation error of the NN in real time. Compared with other inversion-based methods, the NN is directly used as the hysteresis compensator without needing inversion. Additionally, the proposed method does not require pre-training of the NN since the weights can be dynamically updated. To verify the effectiveness and robustness of the proposed method, a series of experiments have been conducted on the self-built exoskeleton robot. Compared with other popular control methods, the proposed method can track the desired trajectory faster, and tracking accuracy is gradually improved through iterative learning and updating.

(2024) Fuzzy-Based Control for Multiple Tasks With Human-Robot Interaction, in IEEE Trans. Fuzzy Syst., DOI: 10.1109/TFUZZ.2024.3429207.
Keywords: Collaborative robots; Fuzzy logic; fuzzy system control; Fuzzy systems; Impedance; multitask motion planning and control; Multitasking; physical human–robot interaction (pHRI); Robot kinematics; Robots.
Abstract: Driven by the rise of collaborative robots, a lot of work has focused on the transparency and stability of physical human-robot interaction (pHRI), in which most of the efforts do not take the requirement of multiple tasks into account. However, the spectrum of applications for collaborative robots has been continuously broadened, and robots without the ability to perform multiple tasks simultaneously may not be capable of collaborating in certain scenarios. In this article, we provide a fuzzy-based multitask intelligent control framework of collaborative robots for pHRI. Our controller formulation consists of “outer-loop” and “inner-loop.” In the “outer-loop,” a fuzzy logic system predicts human desired motion trajectory for the robot to track. In the “inner-loop,” the robot is driven by a hierarchical multitask controller to track the trajectory generated by the “outer-loop” and perform other subtasks simultaneously. With null space projections, the whole task stack can be implemented in a strict task hierarchy in order of priority. The weight-tuning law of the FLSs and the hierarchical multitask control law are given based on Lyapunov stability analysis. The proposed control framework is applied to a mobile manipulator and the effectiveness is verified by exploratory experiments. Results confirm the effectiveness of the proposed control framework and compare its performance with other approaches.

(2024) Synthesize Personalized Training for Robot-Assisted Upper Limb Rehabilitation With Diversity Enhancement, in IEEE Trans. Vis. Comput. Graph., DOI: 10.1109/TVCG.2023.3308940.
Keywords: Algorithms; Automation; Computer Graphics; End effectors; Exercise planning; Game theory; Games; Humans; multi-objective optimization; Multiple objective analysis; Optimization; Rehabilitation; Rehabilitation robots; Robot kinematics; Robotics - methods; Robots; serious game; Solvers; Task analysis; Training; Trajectory; Upper Extremity; upper limb rehabilitation.
Abstract: For upper limb rehabilitation, the robot-assisted technique in combination with serious games requires well-specified training plans. For the best quality of the rehabilitation process, customized game levels for each user are desired, while it is labor-intensive to design and adjust game levels for different individuals. We work on generating training content for a desktop end-effector rehabilitation robot and propose a method to automatically generate individualized training plans. By modeling the search of the training motions as finding optimal hand paths and trajectories, we introduce solving the design problem with a multi-objective optimization (MO) solver. We further improve the MO solver to enhance the diversity of the solutions. With the proposed approach, our system is capable of automatically generating various training plans considering the training intensity and dexterity of each joint in the upper limb. In addition, the enhanced diversity avoids repeated training plans, which helps motivate the user in the rehabilitation. We test our method with different requirements on the training plans and validate the solutions.;For upper limb rehabilitation, the robot-assisted technique in combination with serious games requires well-specified training plans. For the best quality of the rehabilitation process, customized game levels for each user are desired, while it is labor-intensive to design and adjust game levels for different individuals. We work on generating training content for a desktop end-effector rehabilitation robot and propose a method to automatically generate individualized training plans. By modeling the search of the training motions as finding optimal hand paths and trajectories, we introduce solving the design problem with a multi-objective optimization (MO) solver. We further improve the MO solver to enhance the diversity of the solutions. With the proposed approach, our system is capable of automatically generating various training plans considering the training intensity and dexterity of each joint in the upper limb. In addition, the enhanced diversity avoids repeated training plans, which helps motivate the user in the rehabilitation. We test our method with different requirements on the training plans and validate the solutions.For upper limb rehabilitation, the robot-assisted technique in combination with serious games requires well-specified training plans. For the best quality of the rehabilitation process, customized game levels for each user are desired, while it is labor-intensive to design and adjust game levels for different individuals. We work on generating training content for a desktop end-effector rehabilitation robot and propose a method to automatically generate individualized training plans. By modeling the search of the training motions as finding optimal hand paths and trajectories, we introduce solving the design problem with a multi-objective optimization (MO) solver. We further improve the MO solver to enhance the diversity of the solutions. With the proposed approach, our system is capable of automatically generating various training plans considering the training intensity and dexterity of each joint in the upper limb. In addition, the enhanced diversity avoids repeated training plans, which helps motivate the user in the rehabilitation. We test our method with different requirements on the training plans and validate the solutions.

(2024) A Variable-Admittance Assist-As-Needed Controller for Upper-limb Rehabilitation Exoskeletons, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2024.3398565.
Keywords: Aerospace electronics; Algorithms; Controllers; Elbow (anatomy); Electrical impedance; Exoskeletons; Hand (anatomy); Limbs; Physical Human-Robot Interaction; Prosthetics and Exoskeletons; Redundancy; Rehabilitation; Rehabilitation Robotics; Robot kinematics; Task space; Training; Trajectory; Wrist.
Abstract: Algorithms used for controlling upper limb rehabilitation exoskeletons need to promote active participation of patients in the training while preventing incorrect hand motion and arm posture. Any such framework should also be systematically customizable for individual users. Our work is motivated by the lack of such control frameworks for upper-limb rehabilitation exoskeletons. To address this shortcoming, we have proposed a method that consists of a two-port admittance controller in the task space which regulates the interaction between the exoskeleton and the user in both wrist and upper arm interaction points. Parameters of the virtual model in the wrist interaction port are dynamically adjusted based on both the performance and the intention of the subject, realizing a minimally intervening controller. The upper arm admittance model on the other hand, modulates how strictly reference shoulder-elbow synergies are enforced, making the proposed approach useful for various stages of rehabilitation. To achieve this, interaction forces measured between the exoskeleton and the upper-arm are used to modify the joint reference trajectories (generated by a model-based redundancy resolution strategy) in the null space of wrist motions. This letter presents a complete formulation of the introduced method, proof of the variable-admittance controller’s passivity, and experimental results verifying the feasibility and performance of the proposed controller.;Algorithms used for controlling upper limb rehabilitation exoskeletons need to promote active participation of patients in the training while preventing incorrect hand motion and arm posture. Any such framework should also be systematically customizable for individual users. Our work is motivated by the lack of such control frameworks for upper-limb rehabilitation exoskeletons. To address this shortcoming, we have proposed a method that consists of a two-port admittance controller in the task space which regulates the interaction between the exoskeleton and the user in both wrist and upper arm interaction points. Parameters of the virtual model in the wrist interaction port are dynamically adjusted based on both the performance and the intention of the subject, realizing a minimally intervening controller. The upper arm admittance model on the other hand, modulates how strictly reference shoulder-elbow synergies are enforced, making the proposed approach useful for various stages of rehabilitation. To achieve this, interaction forces measured between the exoskeleton and the upper-arm are used to modify the joint reference trajectories (generated by a model-based redundancy resolution strategy) in the null space of wrist motions.This paper presents a complete formulation of the introduced method, proof of the variable-admittance controller’s passivity, and experimental results verifying the feasibility and performance of the proposed controller.

(2024) Hierarchical Trajectory Deformation Algorithm With Hybrid Controller for Active Lower Limb Rehabilitation, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2024.3396369.
Keywords: Active control; active rehabilitation; Actuation; Algorithms; Assistive robots; Constraints; Controllers; Deformable models; Deformation; Force; hierarchical trajectory deformation algorithm; Human engineering; Human motion; Human-robot interaction; Optimization; Physical human-robot interaction (pHRI); Rehabilitation; rehabilitation robotics; Rehabilitation robots; Robot dynamics; Robots; Smoothness; Trajectories; Trajectory; Velocity distribution.
Abstract: Robot-aided active rehabilitation has shown to be an effective treatment approach for hemiplegic patients. This letter presents an active control framework for lower limb rehabilitation, combining an interaction layer with a hierarchical trajectory deformation algorithm (HTDA), and an assist-as-needed (AAN) layer with a hybrid controller. The HTDA uses constrained optimization in both position and velocity domains to continuously generate smooth reference trajectories based on virtual interaction forces during physical human-robot interaction (pHRI). An additional optimization loop is also implemented to achieve adaptive parameter adjustment for HTDA. Meanwhile, the hybrid controller relies on a force field term and a velocity field term to provide AAN feature. The proposed method is validated on a two-degree-of-freedom lower limb rehabilitation robot for walking with variable step height and step length. The experimental results demonstrate that compared to previously developed admittance model (AM) and trajectory deformation algorithm (TDA), under four different evaluation metrics, HTDA can improve dimensionless squared jerk (DSJ) by 73.6% comparing to AM and improve constraint force percentage (CFP) by 20.4% comparing to TDA. This demonstrate the effectiveness of the proposed framework in reducing human-robot confrontation, especially in improving robot actuation compliance and movement smoothness.

(2024) Optimized Impedance Control of a Lightweight Gait Rehabilitation Exoskeleton Based on Accurate Knee Joint Torque Estimation, in IEEE transactions on medical robotics and bionics, DOI: 10.1109/TMRB.2024.3464671.
Keywords: Actuators; attention mechanism; Attention mechanisms; compliant control; Estimation; Exoskeletons; Gait; Human engineering; Impedance; impedance control; Knee; Knee exoskeleton; Medical robotics; Motors; Neural networks; Parameters; Rehabilitation; Rehabilitation robots; Robot dynamics; Signal generation; Torque; torque estimation.
Abstract: In recent years, with the increasing problem of an aging population, there has been a significant increase in the number of stroke patients presenting with motor dysfunction of the lower limbs. In this study, a knee exoskeleton rehabilitation robot driven by a quasi-direct driver actuator is designed. The torque generation model is constructed based on the TCN-LSTM hybrid neural network, and the knee joint torque is generated by sEMG and angle signal. A joint attention mechanism is introduced to enhance the accuracy of torque generation model. The impedance control parameters are adaptively adjusted in accordance with the joint torque. The experimental results demonstrate that the TCN-LSTM hybrid neural network is capable of effectively estimating torque, the mean MAE and CC of the proposed model are 1.141Nm and 93.7%, respectively. The optimized impedance control can optimize the initial value of the impedance parameter, which reduced the torque error by 5.54% and 50.64% at uphill tasks and walking task, respectively, and adaptively adjust the impedance parameter to ensure the coordination of the gait rehabilitation and the friendly human-robot interaction.

(2024) Adaptive Position Constrained Assist-As-Needed Control for Rehabilitation Robots, in IEEE transactions on industrial electronics (1982), DOI: 10.1109/TIE.2023.3273270.
Keywords: Adaptive control; Assist-as-needed strategy; Closed loops; Constraints; Continuity (mathematics); Control methods; Dynamical systems; Exoskeletons; Feedback control; Muscles; position constraints; Rehabilitation; Rehabilitation robots; robot assistance level metric; Robot control; Robots.
Abstract: In rehabilitation practice, motivating patients with neurological injuries to actively increase muscle activity and ensure their safety are important. Therefore, this study proposed a position-constrained assist-as-needed (AAN) control method for rehabilitation robots. A human–robot interaction system with position constraints was first established based on prescribed performance. Aiming at implementing the AAN strategy, the robot assistance level metric (RALM), a constructed global continuous differentiable function incorporating dead zone and saturation characteristics, was introduced to quantify the robotic assistance and facilitate seamless operation. To bridge the gap between the position constraints and the AAN strategy, a sliding manifold was constructed for the constrained human–robot dynamic system, where RALM was regarded as a weight factor to achieve a human-dominated mode, a robot-dominated mode, and their smooth transition, regarded as a human–robot shared mode. The stability of the closed-loop system was guaranteed by using the Lyapunov theory, and the proposed controller was verified by several physical experiments on a knee exoskeleton driven by pneumatic muscles.;In rehabilitation practice, motivating patients with neurological injuries to actively increase muscle activity and ensure their safety are important. Therefore, this study proposed a position-constrained assist-as-needed (AAN) control method for rehabilitation robots. A human-robot interaction system with position constraints was first established based on prescribed performance. Aiming at implementing the AAN strategy, the robot assistance level metric (RALM), a constructed global continuous differentiable function incorporating dead-zone and saturation characteristics, was introduced to quantify the robotic assistance and facilitate seamless operation. To bridge the gap between the position constraints and the AAN strategy, a sliding manifold was constructed for the constrained human-robot dynamic system, where RALM was regarded as a weight factor to achieve a human-dominated mode, a robot-dominated mode, and their smooth transition, regarded as a human-robot shared mode. The stability of the closed-loop system was guaranteed by using the Lyapunov theory, and the proposed controller was verified by several physical experiments on a knee exoskeleton driven by pneumatic muscles (PMs).

(2024) Multimode Control Strategy for Robotic Rehabilitation on Special Orthogonal Group SO(3), in IEEE transactions on industrial electronics (1982), DOI: 10.1109/TIE.2023.3257378.
Keywords: Active control; Algorithms; Attitude control; Contours; Control methods; Disease control; Euclidean geometry; Euclidean space; Exoskeletons; Feedback linearization; Human motion; Hybrid control; Multimode control; Neural networks; Radial basis function; Rehabilitation; Rehabilitation robots; Robot control; Stability analysis; Topology; Tracking control; Tracking errors; Trajectory control.
Abstract: Robot-assisted rehabilitation for three-degree-of-freedom joints, such as hip and ankle, is significant for patients with motor function injuries. The control of such robots involves attitude control. To adapt to different disease stages, multimode hybrid control is considered to be one of the best choices. Passive mode is based on trajectory tracking control, whereas active mode is based on field-based assist-as-needed (AAN) control. The key to AAN control is the solution of the closest attitude point. However, the attitude point belongs to a special orthogonal group SO(3), and its topology is completely different from Euclidean space, which causes difficulties in the solution. Both passive and active control methods are affected by the inaccuracy of model parameters and external disturbances. Therefore, this article proposes a multimode hybrid control method on SO(3). First, the expressions of trajectory tracking and contour tracking errors are proposed. To solve the contour tracking error, a feedback linearization algorithm based on a sliding surface is used. A radial basis function neural network is used for adaptive compensation. Subsequently, a controller for different modes is designed, and its stability is analyzed. Experiments are conducted using a hip exoskeleton, and the results verify the effectiveness of the proposed control method.

(2024) A Cable-Driven Upper Limb Rehabilitation Robot with Muscle-Synergy-Based Myoelectric Controller, in IEEE Trans. Robot., DOI: 10.1109/TRO.2024.3411849.
Keywords: Cable-driven upper limb rehabilitation robot; continuous myoelectric controller; Controllers; Electromyography; End effectors; Force; generalization; muscle synergy; Muscles; Myoelectricity; Real time; Rehabilitation robots; Robot control; Robots; Task analysis; Three dimensional motion; Three-dimensional displays; three-dimensional space; Tracking; Training; Trajectory analysis.
Abstract: Surface electromyography (sEMG) signal has been used in upper limb rehabilitation robots (ULRR). However, existing ULRR based on myoelectric controllers suffered from limited generalization ability in estimating 3D motion intention. This study proposed a muscle-synergy-inspired approach to enhance the generalization ability of the myoelectric controller of a cable-driven ULRR. Low-dimensional commands were extracted from sEMG signals based on an EMG-to-muscle activation model and non-negative matrix factorization. The extracted commands were used to estimate the 3D human force. Two different trajectory tracking tasks were selected to test the generalization ability. The system was trained based on training sets where participants performed one task. Then the system was tested using testing sets where participants performed the other task. Finally, the system was verified on real-time robotic control experiment. Results showed the proposed controller achieved better force estimating accuracy, better trajectory tracking accuracy and lower interaction force than the myoelectric controller without considering muscle synergies, which meant the proposed controller yielded better generalization performance.;Surface electromyography (sEMG) signal has been used in upper limb rehabilitation robots (ULRR). However, existing ULRR based on myoelectric controllers suffers from limited generalization ability in estimating three-dimensional (3-D) motion intention. This article proposes a muscle-synergy-inspired approach to enhance the generalization ability of the myoelectric controller of a cable-driven ULRR. Low-dimensional commands are extracted from sEMG signals based on an EMG-to-muscle activation model and non-negative matrix factorization. The extracted commands are used to estimate the 3-D human force. Two different trajectory tracking tasks are selected to test the generalization ability. The system is trained based on training sets where participants perform one task. Then the system is tested using testing sets where participants perform the other task. Finally, the system is verified on real-time robotic control experiment. Results show that the proposed controller achieves better force estimating accuracy, better trajectory tracking accuracy, and lower interaction force than the myoelectric controller without considering muscle synergies, which means the proposed controller yields better generalization performance.

(2024) Adaptive Human-Robot Interaction Torque Estimation With High Accuracy and Strong Tracking Ability for a Lower Limb Rehabilitation Robot, in IEEE/ASME transactions on mechatronics, DOI: 10.1109/TMECH.2024.3394491.
Keywords: Accuracy; Adaptive learning; Covariance matrices; human–robot interaction; Kalman filters; Noise; Polynomials; rehabilitation robot; Torque; torque estimation; Training.
Abstract: Accurate acquisition of interactive information is crucial for the effective execution of rehabilitation training. However, due to model and sensor errors, it is difficult to obtain interactive information accurately and quickly. To overcome these challenges, a novel accurate and fast estimation method for the human-robot interaction torques (HRITs) is proposed in this article. First, the HRIT model with order adaptive adjustment ability (HMOAA) is constructed. The polynomial order of HMOAA can be adaptively adjusted based on the partial state estimation, which is more consistent with the dynamic time-varying characteristics of HRIT. Second, the Sage-Husa adaptive strong tracking Kalman filter (SHASTKF) is designed based on the modified Sage-Husa adaptive Kalman filter (SHAKF) and strong tracking Kalman filter (STKF). The SHASTKF can quickly track the abrupt HRIT changes when the subject suddenly exerts active torques in rehabilitation training. Moreover, it also has the ability to recursively estimate the noise characteristics, and can stably complete the HRIT estimation task when the noise characteristics are unknown. Finally, simulations and experiments are conducted to validate the proposed method, and the comparison results demonstrate that the proposed method has good torque estimation precision and fast tracking ability of abrupt changes in HRITs.

(2024) Nonlinear Observer-Based Sliding Mode Control for Robot-Aided Bilateral Human-Compliant Rehabilitation Training of Upper Limb, in IEEE Trans. Autom. Sci. Eng., DOI: 10.1109/TASE.2024.3499329.
Keywords: admittance controller; Human-compliant rehabilitation training; nonlinear observer; sliding mode.
Abstract: Robotic-assisted rehabilitation therapy has been a promising way in improving upper limb motor function. This paper proposes a multi-mode training control method for a bilateral upper limb rehabilitation robotic system, with which human-compliant rehabilitation training can be provided. Firstly, an admittance controller is built to transform the human-robot interaction force to compliant desired trajectory. Then, by integrating with super-twisting algorithm, a nonlinear observer is designed to estimate the lumped disturbance exerted on the driving revolute joint, including the active force applied by human subject, the force of friction, the model uncertainty, et al. To guarantee that the state of position converges to the desired value in real time, a high-order sliding mode controller combined with the disturbance compensation from the observer is proposed. Additionally, based on the aforementioned several methods, multiple bilateral training modes are constructed for patients in different rehabilitation stages. The overall system including the constructed bilateral rehabilitation robotic system and the proposed control method is verified in several experiments, demonstrating the advantage of the controller on interaction compliance with respect to normal method in addition to the capability of multiple rehabilitation training modes. Note to Practitioners -This work is motivated by the patients’ needs of the compliance and comfort during the human-robot interaction in the robot-aided rehabilitation training process. Thus, a nonlinear observer-based sliding mode controller combined with admittance model is proposed in this paper. The developed control method has the following functionalities: (1) Ensuring the compliance of the desired trajectory via the constructed admittance model. (2) Solving the estimation of lumped disturbance exerted on the robotic system based on nonlinear force observer. (3) Ensuring the trajectory tracking with high accuracy under unknown disturbances via sliding mode controller combined with the observer compensation. The controller can be potentially applied in lots of areas: 1) Human-robot compliant collaboration or interaction, e.g., human-robot cooperative manipulation, robotic surgery; 2) Precise motion of robotic arm under unknown disturbance.

(2024) A Human-centered Kinematics Design Optimization of Upper Limb Rehabilitation Exoskeleton Based on Configuration Manifold, in IEEE Open J. Comput. Soc., DOI: 10.1109/OJCS.2024.3465661.
Keywords: Design optimization; Elbow; Exoskeletons; Kinematics; Kinematics analysis; Manifolds; Optimization; Physical human-robot interaction; Rehabilitation exoskeleton; Robot kinematics; Shoulder.
Abstract: Shoulder repetitive training is of paramount importance for rehabilitation of stroke patients with hemiplegia. This paper investigated kinematic structural optimization of an upper limb rehabilitation exoskeleton’s shoulder structure, aiming to cover the range of motion (ROM) of human shoulder, achieve sufficient dexterity, obtain a compact structure, and avoid collisions with the user within the workspace. Based on the concept of configuration manifold, configuration parameters and joint angle parameters were fused, and parameter optimization was transformed into a searching problem in high-dimensional configuration space. Geometric constraints between human and exoskeleton were described parametrically. Upper limb movements were mapped to the exoskeleton’s configuration space to calculate spatial vectors of joints, and determine whether vectors satisfy constraints. The formulated multi-objective optimisation problem was computed by multi-objective particle swarm optimization (MOPSO) algorithm to determine the shoulder configuration parameters. Experimental results demonstrate that functional rehabilitation exoskeleton (FREE) exhibits a wide ROM, excellent dexterity, and can assist users in completing most activities of daily life (ADLs). The design framework proposed in this paper can help designers determining optimal exoskeleton configurations through formulated constraints.

(2024) Repetitive Control of Knee Interaction Torque via a Lower Extremity Exoskeleton for Improved Transparency During Walking, in IEEE transactions on medical robotics and bionics, DOI: 10.1109/TMRB.2024.3464119.
Keywords: Algorithms; Cycle time; Dynamics; Error reduction; Exoskeletons; Fitness equipment; Gait; Hip; Hip joint; human-robot interaction; Joints (anatomy); Knee; Legged locomotion; Repetitive control; Repetitive controllers; Robots; stability analysis; Torque; Torque control; transfer functions; Treadmills; Walking.
Abstract: We developed, implemented, and assessed the performance of two forms of plug-in type repetitive controllers (RC) for enhancing the transparency of a lower extremity exoskeleton that operates to support walking function. One controller is a first order RC (SING) consisting of a single period matched to the self-selected cadence of the participant. The second is a novel ‘parallel’ RC (PARA) which consists of a library of integrated RCs with varying periods, intended to accommodate a wider range of gait cycle times. We assessed the effects of both RCs under free cadence walking (FREE) and when walking with a metronome prescribing a consistent cadence matching the participants’ self-selected value. Both conditions were evaluated both at fixed speed and under user-driven treadmill control (UDT), where the treadmill speed was regulated by the user’s anterior/posterior position on the treadmill. The implementation of RC to the knee joint of the ALEX II exoskeleton lead to a significant reduction in torque error of 10-15% at the knee joint during swing and smaller, non-significant effects at the hip joint. While the PARA RC reduced knee torque error more than the SING RC during the FREE cadence condition, a 15% reduction vs. 10% reduction, the difference between the two controllers was not statistically significant. During the UDT sections of walking conditions, participants increased GS under both the SING and PARA RC types. After controlling for the increase in torque error associated with speed, both the PARA and the SING controller reduced TE at the knee joint during swing relative to baseline by 13% and 14%, respectively, with no significant effects to the hip joint. Our work presents a novel formulation of RC and demonstrates the feasibility of applying RC to a robotic exoskeleton joint to assist walking. Future work should be geared toward improving the gait cycle prediction algorithm and developing robust methods for accounting for impact dynamics.

(2024) Adaptive Safety-Critical Control With Uncertainty Estimation for Human-Robot Collaboration, in IEEE Trans. Autom. Sci. Eng., DOI: 10.1109/TASE.2023.3320873.
Keywords: Collaboration; Collision avoidance; Human-robot interaction; Lyapunov methods; Machine learning; Robots; safe human–robot collaboration; Safety; Safety-critical control; Task analysis; Uncertainty; uncertainty estimation.
Abstract: In advanced manufacturing, strict safety guarantees are required to allow humans and robots to work together in a shared workspace. One of the challenges in this application field is the variety and unpredictability of human behavior, leading to potential dangers for human coworkers. This paper presents a novel control framework by adopting safety-critical control and uncertainty estimation for human-robot collaboration. Additionally, to select the shortest path during collaboration, a novel quadratic penalty method is presented. The innovation of the proposed approach is that the proposed controller will prevent the robot from violating any safety constraints even in cases where humans move accidentally in a collaboration task. This is implemented by the combination of a time-varying integral barrier Lyapunov function (TVIBLF) and an adaptive exponential control barrier function (AECBF) to achieve a flexible mode switch between path tracking and collision avoidance with guaranteed closed-loop system stability. The performance of our approach is demonstrated in simulation studies on a 7-DOF robot manipulator. Additionally, a comparison between the tasks involving static and dynamic targets is provided. Note to Practitioners-This research addresses the need to improve the safety of robots interacting with humans when performing collaborative tasks. Existing safety-critical control (SCC) approaches do not adequately monitor and continuously limit the state of the robot in Cartesian space, which results in a risk of injury to human operators if there is unexpected behavior during collaboration. Additionally, existing SCC approaches only consider system uncertainty for a single task (i.e. path tracking only or collision avoidance only). These problems limit the applicability of SCC techniques to manufacturing cobots. We address these problems by developing a controller that accounts for uncertainty in robot dynamics, guarantees that the robot end-effector remains within a constrained task space, and continuously modifies its motion in real-time to avoid dynamic obstacles that violate this space. We employ a machine learning approach to estimate the unknown uncertainties in real-time, allowing them to be incorporated within the controller design. The designed controller selects the shortest path for collision avoidance at each sample instant in order to minimize the total motion of the robot.

(2024) A Dynamic Planner for Safe and Predictable Human-Robot Collaboration, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2023.3334977.
Keywords: Behavioral sciences; Collaboration; Collision avoidance; Collision dynamics; Cooperation; human-aware motion planning; Human-robot collaboration; Human-robot interaction; Manipulators; Motion planning; Optimal control; optimization and optimal control; Regulations; Robots; Safety; safety in HRI; Service robots; Trajectory; Tubes.
Abstract: The new face of modern industrial scenarios involves shared workspaces where humans and robots work closely together. To ensure safe human-robot collaboration (HRC), regulations have been updated introducing the ISO/TS15066. However, complying with these regulations often leads to inefficient behavior, such as unnecessarily reducing robot speed or unpredictably changing the robot path, which may negatively affect the operator perception of the robot. In this letter an optimal approach to address together these two issues is proposed. Starting from a desired final configuration, the framework plans a collision-free trajectory for the robot. Subsequently, predictability is taken into account and a set of virtual tubes into which the path of the robot can move is built. Lastly, an optimization problem is solved online to ensure that the robot stays within these tubes and the velocities are compliant with the ISO/TS 15066. The proposed approach has been experimentally validated in two different scenarios: one composed by a mobile manipulator, i.e. a UR10e mounted on a Neobotix MPO-500, and one composed by only a collaborative manipulator, i.e. a UR5e.

(2024) Fuzzy Logic-Based Arbitration for Shared Control in Continuous Human-Robot Collaboration, in IEEE Trans. Fuzzy Syst., DOI: 10.1109/TFUZZ.2024.3386822.
Keywords: Algorithms; Arbitration; Collaboration; Cooperation; Effectiveness; Fuzzy logic; human–robot collaboration (HRC); Impedance; Kalman filters; Leadership; Linguistics; Obstacle avoidance; Robot control; Robots; shared control; Switches.
Abstract: In human-robot collaboration (HRC) tasks, the role of the robot should be naturally and smoothly transitioned between the leader and the follower to guarantee task performance. To realize this, the arbitration of the shared control between the human and the robot needs to be properly designed to assign a degree of leadership to the robot. In this article, we propose a fuzzy logic-based arbitration rule with the help of Kalman filter. Based on this rule, the arbitration can be continuously regulated between zero and one according to the interaction force and the velocity of the human-robot collaboration system with human intention taken into account. Besides, the distance between the system and the obstacle and more generally the environment is also served as a fuzzy input, so that the possible interaction with the environment, e.g., obstacle avoidance, can be considered to ensure system safety. Since our proposed arbitration rule is based on a fuzzy logic, it endows the robot with the capability of continuous reasoning without an explicit form. The effectiveness of the proposed algorithm is evaluated by experiments.

(2024) Simultaneously Learning of Motion, Stiffness, and Force From Human Demonstration Based on Riemannian DMP and QP Optimization, in IEEE Trans. Autom. Sci. Eng., DOI: 10.1109/TASE.2024.3469961.
Keywords: Collision avoidance; contact-rich scenarios; Dynamics; Force; Hidden Markov models; human-like motion/stiffness/force control; Impedance; Kinematics; Learning from human demonstration; Manifolds; Optimization; QP optimization; Quaternions; Riemannian-based DMP; Robots.
Abstract: In this paper, we propose a motion, stiffness, and force learning framework based on an extended dynamic movement primitive (DMP) and quadratic programming (QP) optimization. The objective is to learn kinematic and dynamic operational parameters from a one-shot human demonstration, through measurement and estimation of the motion, 3-dimensional (3-D) endpoint stiffness, and applied forces of the human arm during manipulation tasks. To this end, first, the framework features an extended DMP to model the motion, stiffness, and force variations in Cartesian space and 2-D sphere manifold. Second, to account for collected errors and human-robot operation gaps, a QP optimization is applied to fine-tune the desired position of the controller. Finally, we validate the framework through two experiments in real scenarios on the Franka Emika Panda robot. Experimental results show that the robot can not only inherit the variation laws of motion, stiffness, and force in the human demonstration, but also exhibit certain generalization capabilities to other situations. The framework provides a reference for robots learning multiple skills via a one-shot human demonstration, which finds great potential application in human-robot cooperation, contact-rich scenarios, and skillful operations, where the motion, stiffness, and applied forces need to be considered simultaneously. Note to Practitioners -Fast programming in robotics through skill transfer plays a critical role in next-generation robots entering ordinary people’s lives. Existing research focuses more on skill learning at the kinematic level and lacks on the dynamic level, such as stiffness and contact force. The goal of this paper is to propose a novel framework for robots learning of motion, stiffness, and force variations from a one-shot human demonstration, simultaneously. To this end, a Riemannian-based DMP method is employed to model the variation laws of motion, stiffness, and force in Cartesian space and 2-D sphere manifold, respectively. In this way, the learning module needs to be run only once, and the patterns can also be generalized to other targets without repeated robot teaching and additional time-consuming processes. To accurately reproduce the learned skills, a human-like motion/stiffness/force controller combined with QP optimization is investigated. In this paper, rather than identifying real environmental parameters, we directly use interacted forces during the human demonstration to represent environmental effects and employ QP to update the desired position in a limited range to account for collected errors and human-robot operation gaps. Experiments on button pressing and polishing tasks by the Panda robot have achieved very good results. The work of this paper lays a foundation for multiple skills learning from human demonstration (LfHD).

(2024) Muscle-Targeted Robotic Assistive Control Using Musculoskeletal Model of the Lower Limb, in IEEE transactions on systems, man, and cybernetics. Systems, DOI: 10.1109/TSMC.2024.3506495.
Keywords: Aerospace electronics; End effectors; Force; Force measurement; Human–robot interaction; Kinematics; muscle-targeted control; Muscles; musculoskeletal model; parallel robot (PR); Parallel robots; Real-time systems; robotic assistance; Robots; Tracking loops.
Abstract: Conventional assistive and rehabilitative robotic systems often overlook human biomechanics, particularly muscular forces, as they predominantly operate in joint or task space and focus on position and exchanged forces. Similarly, traditional manual rehabilitation techniques employed by physiotherapists struggle to obtain quantitative measurements and make precise modifications to key human variables, resulting in predominantly qualitative methods and outcomes. In response to these limitations, this article introduces an innovative assistive robot controller that operates in the muscular space, targeting specific muscles in the lower limb, and distinguishing itself from existing solutions that focus primarily on joint or task space. A key innovation of our approach is the real-time measurement of muscular forces during dynamic tasks, obtained from a calibrated musculoskeletal model. These measurements enable the establishment of a multistep closed-loop controller, with the outer loop precisely tracking the desired muscular forces. Implemented within a configurable viscous environment, the controller provides a natural response for the user. Experimental evaluations conducted using a parallel robot designed for rehabilitation demonstrate the controller’s efficacy. Incorporating the outer loop reduced the median relative error of the tracked muscular force by nearly 80% and decreased the variability of this error by over 85% compared to a pure viscous environment defined as the baseline. These findings highlight the potential applications of this control framework in areas, such as assistive robotics and precision rehabilitation. By achieving objective measurement and control, the system may enhance rehabilitation outcomes, offering tailored exercises that match the individual needs, capabilities, and engagement of each patient.

(2023) Tracking Control of Cable-Driven Planar Robot Based on Discrete-Time Recurrent Neural Network With Immediate Discretization Method, in IEEE Trans. Ind. Inform., DOI: 10.1109/TII.2022.3210255.
Keywords: Cable-driven planar robot; discrete real-time tracking control; discrete-time recurrent neural network (DTRNN); Discretization; End effectors; immediate discretization method; Industrial engineering; Informatics; Mathematical models; Neural networks; Numerical models; physical experiment; Real time; Real-time systems; Recurrent neural networks; Robot control; Robots; Service robots; Tracking control.
Abstract: In recent years, the cable-driven planar robot has made fruitful achievements in many fields, but the related researches are scarce yet in the industrial engineering field. In this article, as a powerful tool for solving discrete time-varying problems, the discrete-time recurrent neural network (DTRNN) is extended to drive the cable-driven planar robot for discrete real-time tracking control, which is derived by a new immediate discretization method, and thus, is termed as ID-DTRNN model. Specifically, first, we present the physical structure and mathematical model of the cable-driven planar robot. Then, the new ID-DTRNN model is proposed and applied for driving such cable-driven planar robot, which bases on the a different way of construction of the traditional DTRNN model. Through numerical experiments, the feasibility, validity, and physical reliability of the ID-DTRNN model for discrete real-time tracking control of the cable-driven planar robot are fully verified. In addition, in the real world, physical experiments of the cable-driven planar robot are presented, which successfully promote the development of physical application of the ID-DTRNN model, and fill the gap of such model in the industrial engineering field.

(2023) Visual-Impedance-Based Human-Robot Cotransportation With a Tethered Aerial Vehicle, in IEEE Trans. Ind. Inform., DOI: 10.1109/TII.2023.3240582.
Keywords: Aerial vehicles; Autonomous aerial vehicles; Cameras; Force; Human motion; Impedance; impedance control; physical human–robot interaction (pHRI); Robots; Smoothness; Stability analysis; Vehicle dynamics; Visual observation; visual servoing; Visualization.
Abstract: Physical human-robot interaction in the field of aerial vehicles has received more research attention in recent years. In this article, a visual impedance control strategy for human-aerial robot cooperative transportation with a tethered vehicle is presented. Without a positioning system, the aerial vehicle is controlled to follow the human partner by using cable force and visual features of the object as feedback. Furthermore, being aware of human motion is important to improve efficiency and smoothness of the cooperation. Without measuring velocities of the aerial vehicle and the human, we propose to directly estimate the relative velocity of them by a vision-based velocity observer. This estimated velocity is then integrated into a visual impedance scheme. The stability of the system is rigorously proved by Lyapunov analysis and passivity analysis. Indoor experiments where a human participant transports a long bar with a tethered aerial vehicle are conducted. Results of experiments under different human velocities and intentions demonstrate the effectiveness and reliability of the proposed method.

(2023) Adaptive Cooperative Control Strategy for a Wrist Exoskeleton Using Model-Based Joint Impedance Estimation, in IEEE/ASME transactions on mechatronics, DOI: 10.1109/TMECH.2022.3211671.
Keywords: Adaptation models; Adaptive control; Adaptive cooperative control (ACC) strategy; Cooperative control; electromyography (EMG); Estimation; Exoskeletons; Human motion; Impedance; joint impedance; Joints (anatomy); Muscles; musculoskeletal model; Real time; Rehabilitation; Solid modeling; Tracking control; Training; Trajectory control; Wrist; wrist rehabilitation robot.
Abstract: Wrist rehabilitation exoskeletons have gained much attention over the last decades, striving to restore motor functions for patients with neuromuscular disorders. Electromyography signal has been employed to estimate the motion intention to achieve interactive training schemes. However, it is a challenging task to estimate the joint impedance in real time, as it is a crucial parameter for control of exoskeletons. This article proposes an adaptive cooperative control strategy for a wrist exoskeleton based on a real-time joint impedance estimation approach. By explicitly interpreting the underlying transformation in the muscular and skeletal systems, the proposed approach estimates the motion intention and the joint impedance of a human subject simultaneously without additional calibration procedures and regulates the training trajectories and assistance accordingly. Results indicate the proposed method outperforms other training protocols, including the trajectory tracking control and the fixed cooperative control. The proposed control strategy provides an additional 66.25% motion deviation when estimated joint torque increases 12.36%, which enhances the training effectiveness and the interaction safety and promotes subjects’ active engagement.

(2023) Towards Human-Robot Collaborative Surgery: Trajectory and Strategy Learning in Bimanual Peg Transfer, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2023.3285478.
Keywords: Algorithms; Autonomous agents; Collaboration; Human-Robot Interaction; Imitation Learning; Intelligent agents; Machine learning; Maximum entropy; Medical Robots; Robot control; Robotic surgery; Robots; Shared Control; Surgeons; Surgery; Task analysis; Training; Trajectory; Workload; Workloads.
Abstract: While the traditional control of surgical robots relies on fully manual teleoperations, human-robot collaborative systems promise to address issues such as workspace constrains and laborious tasks. In particular, shared control can reduce the surgeon’s workload and improve the overall surgical performance by supporting the surgeon effort during movements while keeping them in charge of complex control phases. In this letter, we propose a segmentation of the bimanual peg transfer task that alternates manual and autonomous control correspondingly. The authority allocation in this shared control framework considers both the limitation of learning-based methods and the higher dexterity of humans during physical interaction. The motion and strategies are transferred from an expert human to a da Vinci Research Kit (dVRK) using an epsilon-greedy on a maximum entropy inverse reinforcement learning algorithm. The model generated enables to train an intelligent agent that can skillfully collaborate with the human operator during the surgical task. The proposed shared control framework is verified both on a virtual platform and then on a real dVRK to assess its usability and robustness. The results show that, compared to traditional teleoperation, our method can achieve faster and more consistent peg transfers. An analysis of the participants’ effort also reveals a significantly lower perception of the workload.;While the traditional control of surgical robots relies on fully manual teleoperations, human-robot collaborative systems promise to address issues such as workspace constrains and laborious tasks. In particular, shared control can reduce the surgeon’s workload and improve the overall surgical performance by supporting the surgeon effort during movements while keeping them in charge of complex control phases. In this letter, we propose a segmentation of the bimanual peg transfer task that alternates manual and autonomous control correspondingly. The authority allocation in this shared control framework considers both the limitation of learning-based methods and the higher dexterity of humans during physical interaction. The motion and strategies are transferred from an expert human to a da Vinci Research Kit (dVRK) [1] using an epsilon-greedy on a maximum entropy inverse reinforcement learning algorithm. The model generated enables to train an intelligent agent that can skillfully collaborate with the human operator during the surgical task. The proposed shared control framework is verified both on a virtual platform and then on a real dVRK to assess its usability and robustness. The results show that, compared to traditional teleoperation, our method can achieve faster and more consistent peg transfers. An analysis of the participants’ effort also reveals a significantly lower perception of the workload.

(2023) Robust Admittance Control with Complementary Passivity, in IEEE Control Syst. Lett., DOI: 10.1109/LCSYS.2023.3286812.
Keywords: Admittance control; Force; Human-robot interaction; Linear systems; Multi-objective complementary control; Q measurement; Robots; Robust admittance control; Robustness.
Abstract: This paper studies a robust admittance control problem with a passivity requirement for stable and unstable linear time-invariant systems, motivated by control issues originated from physical human-robot interaction. A complementary admittance control structure is proposed and analyzed, revealing that the nominal performance (admittance tracking and passivity) is decoupled from robustness. Simulations on the admittance control for human arm strength augmentation with a passivity requirement validate the proposed controller design.

(2023) Set-Membership Adaptive Robot Control With Deterministically Bounded Learning Gains, in IEEE Trans. Ind. Inform., DOI: 10.1109/TII.2022.3220892.
Keywords: Adaptive algorithms; Adaptive control; Adaptive robot control; Algorithms; Asymptotic methods; Closed loops; Dynamic stability; Ellipsoids; Feedback control; Heuristic algorithms; Informatics; Lower bounds; neural network (NN) approximation; Neural networks; parameter identification; Prediction algorithms; Robot arms; Robot control; Robot dynamics; robot manipulator; Robots; System dynamics; Torque.
Abstract: As a powerful set-membership adaptive identification algorithm, the optimal bounded ellipsoid (OBE) enables fast convergence speeds because it exploits a priori information about system dynamics by estimating sets of feasible solutions rather than single-point solutions. However, its learning gain matrix suffers from vanishing or unbounded growth, which seriously limits its practical performance. In this article, a novel OBE algorithm is proposed to ensure that the learning gain matrix is constrained by upper and lower bounds, which are unaffected by the hardly predictable excitation levels and can be determined before implementing the algorithm. Thus, the system robustness and tracking capability for time-varying dynamics can be improved. In light of the proposed OBE identification algorithm, an adaptive robot control strategy is further proposed, where the robot dynamics are reconstructed through neural networks. The practical partial asymptotic stability of the closed-loop system is demonstrated using the Lyapunov method. Furthermore, noisy acceleration signals and the inversion of the inertial matrix are eliminated with the proposed control strategy. Experimental results on a robot manipulator validate the effectiveness of the proposed approach.

(2023) Combined Admittance Control With Type II Singularity Evasion for Parallel Robots Using Dynamic Movement Primitives, in IEEE Trans. Robot., DOI: 10.1109/TRO.2023.3238136.
Keywords: Behavioral sciences; Control systems design; Controllers; Couplings; Damping; Dynamic movement primitives (DMPs); Electrical impedance; Force; force control; Human motion; Motion perception; parallel robot (PR); Rehabilitation; rehabilitation robotics; Robot control; Singularities; singularity avoidance; Task analysis; Trajectory; Trajectory control.
Abstract: This article addresses a new way of generating compliant trajectories for control using movement primitives to allow physical human-robot interaction where parallel robots (PRs) are involved. PRs are suitable for tasks requiring precision and performance because of their robust behavior. However, two fundamental issues must be resolved to ensure safe operation: first, the force exerted on the human must be controlled and limited, and second, Type II singularities should be avoided to keep complete control of the robot. We offer a unified solution under the dynamic movement primitives (DMP) framework to tackle both tasks simultaneously. DMPs are used to get an abstract representation for movement generation and are involved in broad areas, such as imitation learning and movement recognition. For force control, we design an admittance controller intrinsically defined within the DMP structure, and subsequently, the Type II singularity evasion layer is added to the system. Both the admittance controller and the evader exploit the dynamic behavior of the DMP and its properties related to invariance and temporal coupling, and the whole system is deployed in a real PR meant for knee rehabilitation. The results show the capability of the system to perform safe rehabilitation exercises.

(2023) Nonlinear Subsystem-based Adaptive Impedance Control of Physical Human-Robot-Environment Interaction in Contact-rich Tasks, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2023.3302616.
Keywords: Adaptive Control; Compliance and Impedance Control; Control methods; Control systems design; Controllers; Degrees of freedom; Exoskeletons; Haptic interfaces; Haptics; Haptics and Haptic Interfaces; Human-robot interaction; Impedance; Physical Human-Robot Interaction; Robot control; Robots; Stability criteria; Subsystems; Task analysis; Virtual environments.
Abstract: Haptic upper limb exoskeletons are robots that assist human operators during task execution while having the ability to render virtual or remote environments. Therefore, ensuring the stability of such robots in physical human-robot-environment interaction (pHREI) is crucial. Having a wide range of Z-width, which indicates the region of passively renderable impedance by a haptic display, is also important for rendering a broad range of virtual environments. To address these issues, this study designs subsystem-based adaptive impedance control to achieve a stable pHREI for 7 degrees of freedom haptic exoskeleton. The presented controller decomposes the entire system into subsystems and designs the controller at the subsystem level. The stability of the controller in the presence of contact with a virtual environment and human arm force is proven by employing the concept of virtual stability. Additionally, the Z-width of the 7-DoF haptic exoskeleton is illustrated using experimental data and improved by exploiting varying virtual mass element. Experimental results are provided to demonstrate the performance of the controller. The control results are also compared to state-of-the-art control methods, highlighting the excellence of the designed controller.

(2023) An Active Strategy for Safe Human-Robot Interaction Based on Visual-Tactile Perception, in IEEE Syst. J., DOI: 10.1109/JSYST.2023.3317328.
Keywords: Active safety; Collision avoidance; Collisions; compliance control; Contact force; Control theory; Deceleration; event-driven mechanism; Human-robot interaction; Humanoid; Humanoid robots; human–robot interaction (HRI); Impact loads; Robot sensing systems; Robots; Safety; Tactile discrimination; Visual perception; Visualization; visual–tactile perception.
Abstract: Ensuring safety is crucial for robots collaborating with humans in a shared workspace. Current human-robot interaction (HRI) safety strategies focus on either precollision motion replanning or postcollision force control, using unimodal perception. In this article, we present an active safety strategy that combines precollision and postcollision strategies using a hierarchical control architecture, making it possible to ensure safety throughout the entire HRI process. The inner loop utilizes a whole-body compliance controller to reduce multicontact forces throughout the robot’s body. The outer loop involves velocity scaling that anticipates collisions based on visual perception and decelerates the robot’s velocity before a collision occurs. This deceleration also simplifies the inner force control law to a computationally efficient linear model. Furthermore, to relieve the communication load of large-scale tactile systems and enable real-time control, we propose the vision assisted event-driven mechanism. It selectively transmits tactile information from potential collision areas. We have constructed a humanoid robotic platform called the visual-tactile humanoid robot and conducted four experiments. Our results demonstrate that our proposed strategy actively ensures safety in the whole process of HRI. The comparison study also confirms that the strategy can both reduce the impact forces at the beginning of the collisions and dissipate the subsequent continuous contact forces compared with existing works. This active strategy promises safety for various scenarios where humans and robots require close and frequent physical contacts.

(2023) Spatial Repetitive Impedance Learning Control for Robot-Assisted Rehabilitation, in IEEE/ASME transactions on mechatronics, DOI: 10.1109/TMECH.2022.3221931.
Keywords: Adaptive control; Control systems design; Error analysis; Exoskeletons; Impedance; impedance learning; iterative learning control; Iterative methods; Learning; Legged locomotion; Parameter modification; Parameter uncertainty; Regulation; Rehabilitation; rehabilitation robot; Rehabilitation robots; repetitive learning control; Robot control; Robot dynamics; Robots; spatial periodicity; Task analysis; Time-domain analysis; Tracking errors.
Abstract: In robot-assisted rehabilitation and leg exoskeletons, humans and robots are required to collaboratively complete repetitive tasks with fixed periodic paths. In such applications, impedance learning control can provide variable impedance regulation for improving the performance of physical interactions; however, designing such control is highly challenging owing to the difficulty in modeling human time-varying dynamics. By exploiting the spatial periodicity characteristics of the desired trajectory and human impedance, we propose a novel spatial repetitive impedance learning control strategy to enhance interaction performance. First, a defined spatial operator serves as the mathematics foundation for constructing the robot dynamics in the spatial domain. Then, a spatial impedance learning controller is designed. In this article, time-varying impedance profiles are estimated using spatial full-saturation iterative learning laws, while robotic parameter uncertainties are estimated using the differential adaptation law with projection modification. We validate the uniform convergence of the tracking error through a Lyapunov-like analysis and demonstrate the control effectiveness using an illustrative example. Compared with related results on temporal repetitive learning control, the proposed control approach can enable human-robot system to complete a repetitive task with unspecified speeds according to the users’ strengths and motion capacity.

(2023) Novel Design and Control of a Crank-Slider Series Elastic Actuated Knee Exoskeleton for Compliant Human-Robot Interaction, in IEEE/ASME transactions on mechatronics, DOI: 10.1109/TMECH.2022.3204921.
Keywords: Actuators; Adaptive control; Biomimetic actuators and sensors; Control theory; Controllers; design methodology for mechatronics; Exoskeletons; Fuzzy control; Fuzzy logic; human–robot interaction; Impedance; Knee; medical and rehabilitation robotics; Modulus of elasticity; series elastic actuator (SEA); Sliding mode control; Springs; Springs (elastic); Stiffness; Thigh; Torque.
Abstract: The lower-limb assist exoskeleton plays the role of torque assiting and compliant tracking for wearers to perform tasks. Accurate torque generation, backdrivability performance, low output impedance, and hardware compactness are essential factors for lower-limb exoskeleton to achieve better compliant physical interaction. This research studies a crank-slider series elastic actuator (CS-SEA) that can be used as a compact exoskeleton joint module. The device has a unique crank slider mechanism, and a set of linear springs are equipped inside the slider to guarantee the nonlinear stiffness of its physical impedance so that the torque effect can be improved and a high level of transparency can be achieved. The RBF-based sliding mode controller is chosen as the output torque controller of the exoskeleton, and the adaptive neuro-fuzzy sliding mode control law is designed and its stability is verified. The precise output force control performance of CS-SEA is verified by experiments. The actuator is incorporated into a knee exoskeleton prototype and was worn by the subjects. The experimental results demonstrate the precision of the compliant transparent and torque assisting control while interacting with the human wearer.

(2023) Physical Human-Robot Interaction Control of Variable Stiffness Exoskeleton With sEMG-Based Torque Estimation, in IEEE Trans. Ind. Inform., DOI: 10.1109/TII.2023.3240749.
Keywords: Actuators; Algorithms; Assist-as-needed (AAN); Control algorithms; Estimation; Exoskeletons; Impedance; Joints (anatomy); Mechanical properties; physical human–robot interaction (pHRI); Pulleys; rehabilitation exoskeleton; Robot control; Robots; Springs; Stiffness; surface electromyography signal (sEMG); Torque; Torque sensors (robotics); variable stiffness actuator (VSA).
Abstract: Robotic exoskeleton assistance is an effective method for the treatment of patients with movement disorders. First, this article presents a variable stiffness knee exoskeleton robot, which can independently control stiffness and position and has a large stiffness range under low preload. Then, combined with the designed variable stiffness exoskeleton, a physical human-robot interaction (pHRI) control scheme based on joint torque estimation is proposed. Different from previous studies, this method uses the mechanical properties of the exoskeleton to achieve pHRI without complex control algorithms and force/torque sensors. Furthermore, the joint torque is estimated based on the surface electromyography signal and the Hill-based muscle model, and the exoskeleton can realize assist-as-needed function: when the subject trains with less effort, the exoskeleton maintains a high output physical impedance to provide high tracking accuracy; and when the subject trains with greater effort, the exoskeleton maintains a low output impedance to provide high physical compliance. Finally, we conducted experiments on three healthy subjects and two subjects with lower limb motor dysfunction to verify the effectiveness of the torque estimation method and the pHRI control scheme.

(2023) Data-Driven Adaptive Iterative Learning Control of a Compliant Rehabilitation Robot for Repetitive Ankle Training, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2022.3229570.
Keywords: Actuators; Adaptive control; Ankle rehabilitation robot; Assistive robots; Convergence; Data models; Dynamic models; Human engineering; iterative learning control; Iterative methods; Learning; Muscles; pneumatic muscle; Read only memory; Rehabilitation robots; Robots; Tracking errors; Training.
Abstract: This letter investigates the repetitive range of motion (ROM) training control for a compliant ankle rehabilitation robot (CARR). The CARR utilizes four pneumatic muscle (PM) actuators to manipulate the ankle with three rational degree-of-freedoms (DoFs) and soft human-robot interaction, but the strong-nonlinearity of the PM actuator makes precise tracking difficult. To improve the training effectiveness, a data-driven adaptive iterative learning controller (DDAILC) is proposed based on compact form dynamic linearization (CFDL) with estimated pseudo-partial derivative (PPD). Instead of using a PM dynamic model, the estimated PPD is updated merely by online input-output (I/O) measures. Sufficient conditions are established to guarantee the convergence of tracking errors and the boundedness of control input. Experimental studies are conducted on ten human participants with two therapist-resembled trajectories. Compared with other data-driven methods, the proposed DDAILC demonstrates significant improvement on tracking performance.

(2023) Disturbance Rejection Speed Control Based on Linear Extended State Observer for Isokinetic Muscle Strength Training System, in IEEE Trans. Autom. Sci. Eng., DOI: 10.1109/TASE.2022.3190210.
Keywords: Control equipment; Control systems; Controllers; Damping; Disturbance rejection control; Dynamics; Elbow (anatomy); Error feedback; Feedback control; Force; isokinetic muscle strength training system; Joints (anatomy); linear extended state observer; Muscle strength; Muscles; Nonlinear control; Observers; Performance enhancement; Physical training; Planning; Rehabilitation; Rejection; Speed control; speed planning; Sports training; State observers; Strategy; Strength training; Tracking; Training; Velocity control.
Abstract: Isokinetic muscle strength training is a type of rehabilitation therapy that can effectively improve the motor dysfunction of limbs and enhance joint muscle strength. In active isokinetic training, how to plan the speed reasonably and improve the performance of speed control to fit individualized needs is the core problem of isokinetic muscle strength training equipment control. In this study, a control strategy for isokinetic muscle strength training equipment is proposed, which identifies the user’s motion intention and performs speed planning according to the user’s motion participation so that the equipment produces a dynamic damping effect. This ensures active interaction between the user and the equipment and meets the force-speed requirements of elbow joint muscle training. To handle the uncertain disturbance in the training process, a linear extended state observer is used to estimate the disturbance term, and a tracking differentiator and nonlinear error feedback control are used to improve the speed tracking performance. The experimental results show that the proposed control scheme realizes effective speed control in isokinetic muscle strength training. Note to Practitioners -The bottleneck of isokinetic muscle strength training equipment control system lies in how to plan the speed according to the patient’s movement intention reasonably, and how to improve the performance of speed control in the process of rehabilitation training. This paper designs a speed auto disturbance rejection control strategy for isokinetic muscle strength training system with external interference and model uncertainty in practical application. The experimental results show that the performance of the controller is better than the traditional PI controller, and it can realize isokinetic muscle training. This study provides a method for practitioners interested in developing control systems for isokinetic rehabilitation training equipment.

(2023) An Intelligent Rehabilitation Robot With Passive and Active Direct Switching Training: Improving Intelligence and Security of Human-Robot Interaction Systems, in IEEE Robot. Autom. Mag., DOI: 10.1109/MRA.2022.3228490.
Keywords: Algorithms; Assistive robots; Control algorithms; Control theory; Intelligence; Legged locomotion; Medical robotics; Rehabilitation; Rehabilitation robots; Robot control; Robot dynamics; Robot kinematics; Robots; Safety; Security; Switches; Switching; Training; Velocity; Walking.
Abstract: The demand for walking training robots has increased owing to a serious shortage of rehabilitation physiotherapists. However, the intelligence level of existing rehabilitation training robots is low; these robots cannot realize a direct switch between passive and active training, which makes rehabilitees lack interest in training and deteriorates the effect of rehabilitation training. In this study, a rehabilitation gait training robot is developed in line with the characteristics of human omnidirectional walking. The proposed robot uses a suitable control algorithm to accurately follow the exercise programs prescribed by physical therapists and can realize fine practice results. The novelty of the robot is that passive and active training can be directly and gently switched during walking. The passive training stage has a velocity restriction safety function, whereas the active training stage has a velocity decision function. The purpose of this design is to avoid sudden changes in the robot velocity during the passive training stage and to guarantee the coordination of the human–robot velocity in the active training stage. Comparative simulation analyses and experimental results show that the proposed passive and active direct switching training improves the intelligence and security of rehabilitation robot.

(2023) A Task Performance-Based sEMG-Driven Variable Stiffness Control Strategy for Upper Limb Bilateral Rehabilitation System, in IEEE/ASME transactions on mechatronics, DOI: 10.1109/TMECH.2022.3208610.
Keywords: Actuators; Bilateral rehabilitation; compliant physical human-robot interaction (pHRI); Control theory; Controllers; Elbow; Electromyography; electromyography (EMG); Human engineering; Limbs; Muscles; Rehabilitation; Rehabilitation robots; Robot kinematics; Robotics; Robots; Stiffness; Task analysis; task performance index (TPI); Torque; Training; variable stiffness actuator (VSA).
Abstract: Bilateral rehabilitation robotics can allow hemiplegia patients to regain the cooperative capabilities of both arms by synchronized coordination movements. Furthermore, the variable stiffness actuators (VSA) integrated robotics can offer compliant advantages for human-robot interaction. Although various studies have proposed to improve training safety and comfortability by VSA, few studies have focused on inducing patient active participation by VSA-based variable stiffness control for bilateral rehabilitation. In this article, an surface electromyography (sEMG) driven variable stiffness control framework with a novel training task quantitative factor TPI was proposed to promote patient active participation in upper limb bilateral rehabilitation. The proposed control law integrates an sEMG-driven musculoskeletal model for providing real-time dynamic reference stiffness from the nonparetic limb as a task skill learning guide to the affected limb. Furthermore, the proposed TPI is designed in the high-level controller for rendering smooth and automatic transition among three patient-robot interaction modes for inducing active participants. In the low-level controller, a position-based bilateral impedance control and a cascaded backstepping position control were implemented for compliant task position planning and tracking. Preliminary experimental results show that the proposed method can promote patient active participation by providing minimal intervention assistance for facilitating efficient upper limb rehabilitation.

(2023) Drivable Space of Rehabilitation Robot for Physical Human-Robot Interaction: Definition and an Expanding Method, in IEEE Trans. Robot., DOI: 10.1109/TRO.2022.3189231.
Keywords: Adaptation models; Adaptive learning; Assistive robots; Biological system modeling; Dynamic models; dynamics modeling; Errors; Exoskeletons; Gaussian process; Human performance; physical human–robot interaction (pHRI); Rehabilitation; rehabilitation robot; Rehabilitation robots; Robot kinematics; Robots; Torque; Training.
Abstract: Physical human-robot interaction performance of present rehabilitation robots are still not satisfactory in the clinical practice. Especially, the work space where the robot can be driven smoothly by users is still very limited, which prevents rehabilitation robots from being applied successfully. In this study, a new concept of drivable space is proposed to evaluate the work spaces of rehabilitation robots, and a method for expanding the drivable space is designed based on the dynamics of the coupled human-robot system and human joint characteristics. First, the definition of drivable space is presented based on comparison of human joint torques, and the minimal torques necessary to drive robot joints, which is mainly determined by the torque estimation errors for general rehabilitation robots driven smoothly by motors. Therefore, a method for improving torque estimation accuracies based on dynamics modeling is then designed. A data-driven error prediction method based on Gaussian process regression is proposed to adaptively compensate the model errors, by which the most accurate dynamic model so far for the coupled system can be obtained, and a method for generation of the training dataset, which is used in error prediction, is designed as well. Moreover, the torque-angle relationship of human joints is modeled and used to optimize the torque error distribution, by which it can be proven that the drivable space can be further expanded. Finally, performance of the proposed methods are demonstrated and validated by experiments carried out on a lower limb rehabilitation robot.

(2023) Voluntary Assist-as-Needed Controller for an Ankle Power-Assist Rehabilitation Robot, in IEEE. Trans. Biomed. Eng., DOI: 10.1109/TBME.2022.3228070.
Keywords: Active participation; Adaptation models; Ankle; Ankle Joint; ankle rehabilitation robot; Assistive robots; Control systems; Controllers; Electromyography; Human motion; human-robot cooperation; Humans; Joints (anatomy); Lower Extremity; Participation; performance adaptive; Rehabilitation; Rehabilitation robots; Robot control; Robotics - methods; Robots; Stroke; Stroke Rehabilitation; Torque; Tracking errors; Trajectories; Trajectory tracking; voluntary torque.
Abstract: OBJECTIVEAlthough existing assist-as-needed (AAN) controllers have been designed to adapt the robotic assistance to patients’ movement performance, they ignore patient’s active participation. This study proposed a voluntary AAN (VAAN) controller considering both movement performance and active participation for an ankle rehabilitation robot. METHODSAccording to the trajectory tracking error of the human-robot cooperation movement, the controller can switch among four working modes, including robot-resist, free, robot-assist, and robot-dominant mode. In order to reflect patients’ active participation, the voluntary torque of the ankle joint was estimated by an EMG-driven musculoskeletal model. The control torque in robot-resist, free, and robot-assist mode was determined by the voluntary torque of ankle joint multiplied by an assistance ratio to encourage subjects’ active participation, and a stiff torque was provided in robot-dominant mode. The controller was evaluated with 2 healthy subjects and 5 stroke patients on an ankle rehabilitation robot to investigate the clinical impact on the stroke patients. RESULTSThe experiment results showed that as patients’ disability level increased, the trajectory tracking error increased and the proportion of human-dominant time and the voluntary torque of ankle joint decreased. Moreover, the results showed that the proposed VAAN controller achieved higher human contribution ratio than that of previous studies. CONCLUSIONThe proposed VAAN controller can adapt the working mode to the movement performance and promote the subjects to participate actively. SIGNIFICANCEBased on its performance, the proposed VAAN controller has potential for use in robot-assisted rehabilitation.;Objective: Although existing assist-as-needed (AAN) controllers have been designed to adapt the robotic assistance to patients’ movement performance, they ignore patient’s active participation. This study proposed a voluntary AAN (VAAN) controller considering both movement performance and active participation for an ankle rehabilitation robot. Methods: According to the trajectory tracking error of the human-robot cooperation movement, the controller can switch among four working modes, including robot-resist, free, robot-assist, and robot-dominant mode. In order to reflect patients’ active participation, the voluntary torque of the ankle joint was estimated by an EMG-driven musculoskeletal model. The control torque in robot-resist, free, and robot-assist mode was determined by the voluntary torque of ankle joint multiplied by an assistance ratio to encourage subjects’ active participation, and a stiff torque was provided in robot-dominant mode. The controller was evaluated with 2 healthy subjects and 5 stroke patients on an ankle rehabilitation robot to investigate the clinical impact on the stroke patients. Results: The experiment results showed that as patients’ disability level increased, the trajectory tracking error increased and the proportion of human-dominant time and the voluntary torque of ankle joint decreased. Moreover, the results showed that the proposed VAAN controller achieved higher human contribution ratio than that of previous studies. Conclusion: The proposed VAAN controller can adapt the working mode to the movement performance and promote the subjects to participate actively. Significance: Based on its performance, the proposed VAAN controller has potential for use in robot-assisted rehabilitation.;Although existing assist-as-needed (AAN) controllers have been designed to adapt the robotic assistance to patients’ movement performance, they ignore patient’s active participation. This study proposed a voluntary AAN (VAAN) controller considering both movement performance and active participation for an ankle rehabilitation robot. According to the trajectory tracking error of the human-robot cooperation movement, the controller can switch among four working modes, including robot-resist, free, robot-assist, and robot-dominant mode. In order to reflect patients’ active participation, the voluntary torque of the ankle joint was estimated by an EMG-driven musculoskeletal model. The control torque in robot-resist, free, and robot-assist mode was determined by the voluntary torque of ankle joint multiplied by an assistance ratio to encourage subjects’ active participation, and a stiff torque was provided in robot-dominant mode. The controller was evaluated with 2 healthy subjects and 5 stroke patients on an ankle rehabilitation robot to investigate the clinical impact on the stroke patients. The experiment results showed that as patients’ disability level increased, the trajectory tracking error increased and the proportion of human-dominant time and the voluntary torque of ankle joint decreased. Moreover, the results showed that the proposed VAAN controller achieved higher human contribution ratio than that of previous studies. The proposed VAAN controller can adapt the working mode to the movement performance and promote the subjects to participate actively. Based on its performance, the proposed VAAN controller has potential for use in robot-assisted rehabilitation.

(2023) Spatiotemporal Compliance Control for a Wearable Lower Limb Rehabilitation Robot, in IEEE. Trans. Biomed. Eng., DOI: 10.1109/TBME.2022.3230784.
Keywords: Adaptive control; Compliance; Compliance control; Controllers; Exercise Therapy - instrumentation; Exercise Therapy - methods; Gait; Gait - physiology; Generators; Human engineering; Humans; Legged locomotion; Lower Extremity; Motor task performance; motor variability; Neural networks; Neural Networks, Computer; Passive control; physical human–robot interaction; Radial basis function; Rehabilitation; Rehabilitation robots; Robot control; Robot dynamics; Robot kinematics; Robotics - methods; Robots; Spatiotemporal phenomena; Torque; Trajectories; Trajectory; Walking; Wearable Electronic Devices; wearable lower limb rehabilitation robot; Wearable technology.
Abstract: Compliance control is crucial for physical human-robot interaction, which can enhance the safety and comfort of robot-assisted rehabilitation. In this study, we designed a spatiotemporal compliance control strategy for a new self-designed wearable lower limb rehabilitation robot (WLLRR), allowing the users to regulate the spatiotemporal characteristics of their motion. The high-level trajectory planner consists of a trajectory generator, an interaction torque estimator, and a gait speed adaptive regulator, which can provide spatial and temporal compliance for the WLLRR. A radial basis function neural network adaptive controller is adopted as the low-level position controller. Over-ground walking experiments with passive control, spatial compliance control, and spatiotemporal compliance control strategies were conducted on five healthy participants, respectively. The results demonstrated that the spatiotemporal compliance control strategy allows participants to adjust reference trajectory through physical human-robot interaction, and can adaptively modify gait speed according to participants’ motor performance. It was found that the spatiotemporal compliance control strategy could provide greater enhancement of motor variability and reduction of interaction torque than other tested control strategies. Therefore, the spatiotemporal compliance control strategy has great potential in robot-assisted rehabilitation training and other fields involving physical human-robot interaction.

(2023) Human-Robot Interaction Evaluation-Based AAN Control for Upper Limb Rehabilitation Robots Driven by Series Elastic Actuators, in IEEE Trans. Robot., DOI: 10.1109/TRO.2023.3286073.
Keywords: Actuators; Algorithms; Assist-as-needed (AAN); Control systems design; Dynamics; Force; Human performance; human–robot interaction; Impedance; impedance adaption; Machine learning; Modulus of elasticity; Perturbation theory; Rehabilitation; Rehabilitation robots; Robot control; Robot kinematics; Robots; Sea measurements; series elastic actuator (SEA)-driven robot; Singular perturbation; Stability analysis; Task analysis.
Abstract: Series elastic actuators (SEAs) have been the most popular compliant actuators as they possess a variety of advantages, such as high compliance, good backdrivability, and tolerance to shocks. They have been adopted by various rehabilitation robots to provide appropriate assistance with suitable compliance during human–robot interaction. For a multijoint SEA-driven rehabilitation robot, a big challenge is to develop an assist-as-needed (AAN) method without losing stability during uncertain physical human–robot interaction. For this purpose, this article proposes a human–robot interaction evaluation-based AAN method for upper limb rehabilitation robots driven by SEAs. First, in order to stabilize the SEA-level dynamics, singular perturbation theory is adopted to design a fast time-scale controller. Second, for the robot-level dynamics, an iterative learning algorithm is adopted for impedance adaption according to the task performance and human intention. The interaction force feedback is introduced for human–robot interaction evaluation, and the intensity of robotic assistance will be adjusted periodically according to the evaluation results. The stability of human–robot interaction is provided with the Lyapunov method. Finally, the proposed rehabilitation method is constructed and implemented on a two-degree-of-freedom SEA-driven robot. It handles the uncertain interaction in such a principle that correct movements will lead to less assistance for encouraging participation and incorrect movements will lead to more assistance for effective training. The proposed method adapts to the subject’s intention and encourages higher participation by decreasing impedance learning strength and increasing allowable motion error. It can fit the participants with different motor capabilities and provide adaptive assistance when a specific trainee tries to change his/her participation during rehabilitation. The performance of the AAN method was validated with experimental studies involving healthy subjects.

(2023) AGREE: A Compliant-Controlled Upper-Limb Exoskeleton for Physical Rehabilitation of Neurological Patients, in IEEE transactions on medical robotics and bionics, DOI: 10.1109/TMRB.2023.3239888.
Keywords: Actuation; Antigravity; Arms; compliant control; Control methods; Control systems; Elbow; Elbow (anatomy); Exoskeletons; Human engineering; Impedance; Limbs; physical human-robot interaction (pHRI); Rehabilitation; robotic rehabilitation; Robots; Shoulder; Torque; Torque sensors (robotics); Upper-limb exoskeleton.
Abstract: In this work, we introduce the AGREE exoskeleton, a robotic device designed to assist in upper-limb physical rehabilitation for post-stroke survivors. We detail the exoskeleton design at the mechatronic, actuation, and control levels. The AGREE exoskeleton features a lightweight and adaptable mechanical design, which can be used with both the right and left arm, supporting three active degrees-of-freedom at the shoulder and one at the elbow. The device embodies a spring-pulley anti-gravity system to minimize torque requirements and has torque sensors on each joint for safe and smooth interaction with the user. The AGREE control system, which employs a loadcell-based impedance control method, offers various modes of human-robot interaction, such as passive-assisted, active-assisted, and active-resistive exercises. Results from our experimental characterization demonstrate that the exoskeleton is capable of both compliant and rigid behavior, providing a wide range of haptic impedance and transparent behavior to both user-generated and therapist-generated forces. Our findings indicate that the AGREE exoskeleton may be a viable option for safely assisting patients with neurological conditions.

(2023) Machine Learning in Robot-Assisted Upper Limb Rehabilitation: A Focused Review, in IEEE Trans. Cogn. Dev. Syst., DOI: 10.1109/TCDS.2021.3098350.
Keywords: Robots; Robot sensing systems; Training; Rehabilitation robotics; Estimation; Exoskeletons; End effectors; Human–robot interaction (HRI); intention recognition; machine learning; quantitative assessment; upper limb rehabilitation.
Abstract: Robot-assisted rehabilitation, which can provide repetitive, intensive, and high-precision physics training, has a positive influence on the motor function recovery of stroke patients. Current robots need to be more intelligent and more reliable in clinical practice. Machine learning algorithms (MLAs) are able to learn from data and predict future unknown conditions, which is of benefit to improve the effectiveness of robot-assisted rehabilitation. In this article, we conduct a focused review on machine learning-based methods for robot-assisted upper limb rehabilitation. First, the current status of upper rehabilitation robots is presented. Then, we outline and analyze the designs and applications of MLAs for upper limb movement intention recognition, human–robot interaction control, and quantitative assessment of motor function. Meanwhile, we discuss the future directions of MLAs-based robotic rehabilitation. This review article provides a summary of MLAs for robotic upper limb rehabilitation and contributes to the design and development of future advanced intelligent medical devices.

(2023) Trajectory Deformation With Constrained Optimization for Bilateral Rehabilitation Robots, in IEEE/ASME transactions on mechatronics, DOI: 10.1109/TMECH.2023.3239616.
Keywords: Active control; Bilateral treatment; constrained optimization; Constraints; Deformation; Exoskeletons; Interactive control; Monitoring; Optimization; physical human–robot interaction (pHRI); Rehabilitation robots; Robot dynamics; Robots; Safety; Strain; Training; Trajectory; Trajectory control; trajectory deformation; Trajectory optimization.
Abstract: Robot-aided bilateral treatment has been verified to be an effective training program for hemiplegic rehabilitation. In this article, a reference-free active control framework based on optimal trajectory deformation is proposed to ensure the safety requirements in the leader-follower paradigm of bilateral treatment. A constrained optimization method is developed to handle the motion constraints, which are constructed by quantitative assessments of typical impairment in stroke patients, such as reduced range of motion, resistance to passive movement, and disturbed quality of movement. Then, by optimally deforming the robotic trajectory, abnormal motion patterns that lead to safety issues can be rectified. Furthermore, the physically interactive trajectory deformation is employed to achieve active control without a predefined trajectory. At last, all approaches are verified on a robotic platform with a 2-DoF lower-limb exoskeleton. Experimental results demonstrate the effectiveness of proposed control scheme in rectifying abnormal motion patterns and enhancing human-robot interaction.

(2023) Event-Triggered Adaptive Hybrid Torque-Position Control (ET-AHTPC) for Robot-Assisted Ankle Rehabilitation, in IEEE transactions on industrial electronics (1982), DOI: 10.1109/TIE.2022.3183358.
Keywords: Adaptive assistance; Adaptive control; adaptive torque control; Ankle; ankle rehabilitation robot; Artificial neural networks; Assistive robots; Control theory; Controllers; event-triggered position control; Force; Indexes; Muscles; Neural networks; Recovery; Rehabilitation; Rehabilitation robots; Robot control; Robots; Torque; Torque control; Tracking control; Tracking errors; Trajectory control.
Abstract: Ankle rehabilitation for an increasing number of strokes is highly demanded, and robot-assisted approach has shown great potential. Since the required movement and force assistances will concurrently change during rehabilitation sessions, the robotic assistances are supposed to be adjusted accordingly. In order to achieve both adaptive torque and synchronous position control for the robot in practice, a novel event-triggered adaptive hybrid torque-position control is proposed in this article for a developed ankle rehabilitation robot driven by pneumatic muscles. In the novel adaptive torque control scheme, the assistive torque adapted to the patient’s recovery state is adjusted by a designed robot-assisted rehabilitation index mapping from the clinical assessment scale. The robotic assistance output is online corrected by patient’s performance, based on a correcting index calculated by interaction torque and tracking errors. Then, a model-based event-triggered optimal position controller is established and a critic neural network is introduced to reduce the control law update frequency for fast trajectory tracking. The stability of the overall system is proved by the Lyapunov theorem. A series of experiments were conducted on the ankle rehabilitation robot to validate the controller’s fast trajectory tracking and adaptive assistance capacity, which can online adjust the robot’s assistive torque and allowable movement range for patients at different recovery stages.

(2023) Stiffness-Observer-Based Adaptive Control of an Intrinsically Compliant Parallel Wrist Rehabilitation Robot, in IEEE T. Hum.-Mach. Syst., DOI: 10.1109/THMS.2022.3211164.
Keywords: Actuation; Actuators; Adaptive control; Biomimetic muscle actuators (BMAs); Biomimetics; Controllers; End effectors; Fuzzy control; fuzzy-logic-based adaptive controller; Muscles; parallel robot; Prototypes; Rehabilitation; Rehabilitation robots; Robot kinematics; Robots; Stiffness; stiffness-observer; Trajectory control; Wrist; wrist rehabilitation robot; wrist stiffness model.
Abstract: Disability from injuries and diseases is a global problem affecting a large population; however, due to a lack of therapists and labor-intensive procedures, only a few benefits from rehabilitation. Robots can assist therapists in treating many patients simultaneously, but the existing solutions need improvements in their mechanism, actuation, and control. This article presents a four-link parallel end-effector robot for wrist joint rehabilitation. The proposed robot employs biomimetic muscle actuators (BMA) that provide intrinsic compliance to the robotic system. A fuzzy-based model is developed to identify the nonlinear nature of BMAs. The stiffness-observer learns subject-specific stiffness, which is used to modify the robot reference trajectories. An adaptive controller uses the fuzzy model and stiffness-observer and simultaneously controls the four BMAs to provide three degrees of rotational freedom to the robot end-effector. The feasibility of the robot mechanism and the controller was evaluated through proof of concept experiments conducted with three unimpaired human subjects. It was found that the controller was able to guide the robot-human system on the commanded trajectories in the presence of parallel actuation of compliant and nonlinear BMAs. Furthermore, the controller was also able to modify the commanded trajectories in the higher stiffness regions of the wrist workspace.

(2023) Performance-Based Iterative Learning Control for Task-Oriented Rehabilitation: A Pilot Study in Robot-Assisted Bilateral Training, in IEEE Trans. Cogn. Dev. Syst., DOI: 10.1109/TCDS.2021.3072096.
Keywords: Admittance; Algorithms; Bilateral upper limb; Control algorithms; Control theory; Convergence; End effectors; Force sensors; Fuzzy logic; Human subjects; Indicators; Interactive control; Iterative methods; Machine learning; Muscular fatigue; Parameters; Performance evaluation; performance-based; Regression models; Rehabilitation; Robot control; Robot learning; robot-assisted rehabilitation; Robots; subject-specific; Task analysis; Training; training task planning.
Abstract: Active participation from human subjects can enhance the effectiveness of robot-assisted rehabilitation. Developing interactive control strategies for customized assistance is therefore essential for encouraging human-robot engagement. However, existing human-robot interactive control strategies lack precise evaluation indicators with effective convergence method to steadily and rapidly customize appropriate assistance during task-oriented training. This study proposes a performance-based iterative learning control algorithm for robot-assisted training, which aims at providing subject-specific robotic assistance to encourage active participation. Three performance indicators based on a Fugl-Meyer assessment (FMA) regression model are introduced to associate clinical scales with robot-based measures, and a fuzzy logic is employed for comprehensive performance evaluation. To increase efficient training time, a piecewise learning rate-based iterative law is applied to quickly converge to a subject-specific control parameter session by session. The proposed strategy is preliminarily estimated for a case of bilateral upper limb training with an end-effector-based robotic system. The experimental results with human subjects indicate that the proposed strategy can obtain appropriate parameters after only several iterations and adapt to random perturbations (like muscle fatigue).

(2023) Cooperative Assist-As-Needed Control for Robotic Rehabilitation: A Two-Player Game Approach, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2023.3261750.
Keywords: Adaptation models; Adaptive control; assist-as-needed control; Autonomy; Cooperative control; Cost function; Game theory; Games; Human performance; Human-robot interaction; Impedance; Iterative methods; non-zero-sum game; Optimal control; Physical human-robot interaction; Rehabilitation; rehabilitation robotics; Robot control; Robots; shared autonomy; Zero sum games.
Abstract: This letter presents an adaptive optimal control strategy for developing assist-as-needed robotic rehabilitation. The primary goal is to encourage patient participation and increase the effectiveness of training sessions by minimizing robot intervention while following a predefined path. To achieve this, the problem is modeled as a two-player non-zero-sum game, with cooperative and individual objectives for the human and robot specified as different cost functions. Policy iteration techniques are adopted for learning the optimal solution online. The shared autonomy feature is specifically achieved through seamless adaptation of the robot’s autonomy according to its estimation of human intention. The performance of the proposed approach is illustrated in several simulations and experimental studies.;This paper presents an adaptive optimal control strategy for developing assist-as-needed robotic rehabilitation. The primary goal is to encourage patient participation and increase the effectiveness of training sessions by minimizing robot intervention while following a predefined path. To achieve this, the problem is modeled as a two-player non-zero-sum game, with cooperative and individual objectives for the human and robot specified as different cost functions. Policy iteration techniques are adopted for learning the optimal solution online. The shared autonomy feature is specifically achieved through seamless adaptation of the robot’s autonomy according to its estimation of human intention. The performance of the proposed approach is illustrated in several simulations and experimental studies.

(2023) Isokinetic Muscle Strength Training Strategy of an Ankle Rehabilitation Robot Based on Adaptive Gain and Cascade PID Control, in IEEE Trans. Cogn. Dev. Syst., DOI: 10.1109/TCDS.2022.3145998.
Keywords: Adaptive algorithms; Adaptive control; Adaptive gain; Ankle; ankle rehabilitation robot; cascade proportion integration differentiation (PID) controller; Controllers; Immune system; motion intention recognition; Muscle strength; Muscles; Physical training; Proportional integral derivative; Rehabilitation; Rehabilitation robotics; Rehabilitation robots; Resistance training; Servomotors; Smoothness; Sports training; Strength training; Torque; Torque measurement; Training; Velocity.
Abstract: Isokinetic muscle strength training refers to the mode of movement based on constant speed and variable resistance, which can guarantee the maximum resistance and force–distance output of each muscle in different angles of exercise. In this article, we develop an isokinetic muscle strength training strategy based on the adaptive gain and cascade proportion integration differentiation (PID) controller for ankle rehabilitation on our newly developed ankle robotic system. First, a heuristic threshold intention recognition method based on time-series integration was proposed to precisely recognize the motion intention, thus inducing the motion in this direction. Then, an adaptive gain algorithm was developed to provide speed gain for the isokinetic training, to avoid jitter during speed switching. Finally, the isokinetic characteristic was realized by the cascade PID controller, which can control the motion velocity in a given range with fast response speed. Experiments with healthy subjects showed good performance in the smoothness of the control system, the accuracy, and the real-time performance of velocity tracking. By introducing the isokinetic characteristic in ankle rehabilitation, the ankle robot can provide resistance training at a constant speed no matter how much force the patient uses, which is a very functional supplement and improvement for ankle rehabilitation.

(2023) Using Repetitive Control to Enhance Force Control During Human-Robot Interaction in Quasi-Periodic Tasks, in IEEE transactions on medical robotics and bionics, DOI: 10.1109/TMRB.2023.3237766.
Keywords: Error analysis; Error reduction; Force; Force control; Human engineering; Human motion; Human-robot interaction; Performance enhancement; Rehabilitation; Repetitive control; Robot control; Robots; Stability analysis; Task analysis; Transfer functions.
Abstract: We investigated the use of repetitive control (RC) to enhance force control during human-robot interaction in quasi-periodic tasks. We first developed a two-mass spring damper model and formulated three different RCs under force control: a nth 1 order RC (RC-1), a nth 3 order RC designed for random period error (RC-RPE), and a nth 3 order RC designed for constant period error (RC-RPE). Then, we quantified the performance of these three RCs through simulations and experiments conducted on a bench top linear platform, subject to nominal cyclical inputs (input signal and RC frequency: 0.5 Hz), and subject to inputs with random and constant period errors. During evaluation, we compared the performance achieved with the RCs with the one achievable with a passive proportional controller (PPC), subject to known theoretical limits for passivity and coupled stability. In nominal simulation conditions, the RC-1 reduced force error most effectively to 0.7% of the error measured with PPC. In real-world nominal experiments the RC-RPE most effectively reduced force error to 17% of the error achieved with the PPC. Subject to inputs with constant period errors, the three RCs performed similarly and better than PPC for period error values below 0.02 Hz, with the RC-CPE performing the best above 0.02 Hz period error. Subject to inputs with random period errors, all RCs performed better than PPC up to 0.09 Hz of period error, with RC-1 significantly outperforming the two other RCs. Our results indicate that RC can be successfully integrated into force control schemes to improve performance beyond the one achievable with a PPC, in the range of period variability expected in applications such as assistance or rehabilitation of cyclical human movements.;We investigated the use of repetitive control (RC) to enhance force control during human-robot interaction in quasiperiodic tasks. We first developed a two-mass spring damper model and formulated three different RCs under force control: a 1st order RC (RC-1), a 3rd order RC designed for random period error (RC-RPE), and a 3rd order RC designed for constant period error (RC-RPE). Then, we quantified the performance of these three RCs through simulations and experiments conducted on a bench top linear platform, subject to nominal cyclical inputs (input signal and RC frequency: 0.5 Hz), and subject to inputs with random and constant period errors. During evaluation, we compared the performance achieved with the RCs with the one achievable with a passive proportional controller (PPC), subject to known theoretical limits for passivity and coupled stability. In nominal simulation conditions, the RC-1 reduced force error most effectively to 0.7% of the error measured with PPC. In realworld nominal experiments the RC-RPE most effectively reduced force error to 17% of the error achieved with the PPC. Subject to inputs with constant period errors, the three RCs performed similarly and better than PPC for period error values below 0.02 Hz, with the RC-CPE performing the best above 0.02 Hz period error. Subject to inputs with random period errors, all RCs performed better than PPC up to 0.09 Hz of period error, with RC-1 significantly outperforming the two other RCs. Our results indicate that RC can be successfully integrated into force control schemes to improve performance beyond the one achievable with a PPC, in the range of period variability expected in applications such as assistance or rehabilitation of cyclical human movements.

(2023) sEMG-Based Adaptive Cooperative Multi-Mode Control of a Soft Elbow Exoskeleton Using Neural Network Compensation, in IEEE Trans. Neural Syst. Rehabil. Eng., DOI: 10.1109/TNSRE.2023.3306201.
Keywords: Training; Exoskeletons; Elbow; Admittance; Torque; Backstepping; Sliding mode control; Soft elbow exoskeleton; adaptive cooperative multi-mode control; sEMG; neural network compensation; active participation.
Abstract: Soft rehabilitation exoskeletons have gained much attention in recent years, striving to assist the paralyzed individuals restore motor functions. However, it is a challenge to promote human-robot interaction property and satisfy personalized training requirements. This article proposes a soft elbow rehabilitation exoskeleton for the multi-mode training of disabled patients. An adaptive cooperative admittance backstepping control strategy combined with surface electromyography (sEMG)-based joint torque estimation and neural network compensation is developed, which can induce the active participation of patients and guarantee the accomplishment and safety of training. The proposed control scheme can be transformed into four rehabilitation training modes to optimize the cooperative training performance. Experimental studies involving four healthy subjects and four paralyzed subjects are carried out. The average root mean square error and peak error in trajectory tracking test are 3.18 degrees and 5.68 degrees . The active cooperation level can be adjusted via admittance model, ranging from 4.51 degrees /Nm to 10.99 degrees /Nm. In cooperative training test, the average training mode value and effort score of healthy subjects (i.e., 1.58 and 1.50) are lower than those of paralyzed subjects (i.e., 2.42 and 3.38), while the average smoothness score and stability score of healthy subjects (i.e., 3.25 and 3.42) are higher than those of paralyzed subjects (i.e., 1.67 and 1.71). The experimental results verify the superiority of proposed control strategy in improving position control performance and satisfying the training requirements of the patients with different hemiplegia degrees and training objectives.

(2023) Choosing Stiffness and Damping for Optimal Impedance Planning, in IEEE Trans. Robot., DOI: 10.1109/TRO.2022.3216078.
Keywords: Algorithms; Cartesian coordinates; Closed loops; Control; Controllers; Damping; End effectors; Exact solutions; Impedance; Inertia; Jacobian matrices; Numerical methods; Optimization; Perturbation; Planning; Robot arms; Robots; Robustness (mathematics); Sensors; Stiffness; Task analysis.
Abstract: The attention given to impedance control in recent years does not match a similar focus on the choice of impedance values that the controller should execute. Current methods are hardly general and often compute fixed controller gains relying on the use of expensive sensors. In this article, we address the problem of online impedance planning for Cartesian impedance controllers that do not assign the closed-loop inertia. We propose an optimization-based algorithm that, given the Cartesian inertia, computes the stiffness and damping gains without relying on force/torque measurements and so that the effects of perturbations are less than a maximum acceptable value. By doing so, we increase robot resilience to unexpected external disturbances while guaranteeing performance and robustness. The algorithm provides an analytical solution in the case of impedance-controlled robots with diagonally dominant inertia matrix. Instead, established numerical methods are employed to deal with the more common case of nondiagonally dominant inertia. Our work attempts to create a general impedance planning framework, which needs no additional hardware and is easily applicable to any robotic system. Through experiments on real robots, including a quadruped and a robotic arm, our method is shown to be employable in real time and to lead to satisfactory behaviors.

(2023) Deep Imitation Learning of Nonlinear Model Predictive Control Laws for a Safe Physical Human-Robot Interaction, in IEEE Trans. Ind. Inform., DOI: 10.1109/TII.2022.3217833.
Keywords: Algorithms; Artificial neural networks; Collision avoidance; Completion time; Control theory; Cost function; Human motion; Industrial robotics; Kinematics; Machine learning; Manipulators; model predictive control (MPC); Motion planning; Neural networks; Nonlinear control; physical human–robot interaction; Predictive control; Robot arms; Robot dynamics; Robots; Safety; Service robots.
Abstract: This article proposes motion planning algorithms for industrial manipulators in the presence of human operators based on deep neural networks (DNNs), aimed at imitating the behavior of a nonlinear model predictive control (NMPC) scheme. The proposed DNN solutions retain the safety features of NMPC in terms of speed and separation monitoring, defined according to the guidelines in the ISO/TS 15066 standard. At the same time, they improve the robot performance in terms of task completion time, and of a posteriori evaluation of the NMPC cost function on experimental data. The reasons for this improvement are the reduced computational delay of running a DNN compared to solving the nonlinear programs associated to NMPC, and the ability to implicitly learn how to predict the human operator’s motion from the training set.

(2023) Adaptive Intention-Driven Variable Impedance Control for Wearable Robots With Compliant Actuators, in IEEE Trans. Control Syst. Technol., DOI: 10.1109/TCST.2022.3222728.
Keywords: Actuators; Adaptation models; Adaptive control; Closed loops; Controllers; Dynamics; Estimation; Exoskeletons; Feedback control; Hidden Markov models; Human motion; Human motion intention; Impedance; Random noise; Robot control; Robot dynamics; Robots; stochastic analysis; variable impedance control; Wearable robots; Wearable technology.
Abstract: Understanding human motion intention is fundamental to wearable robots, which are designed to provide assistance by assessing the intention of wearers. Although modeling human motion intention during physical interaction reveals the fundamental properties of wearable robots, uncertainty and random noise are commonly neglected in existing works. This article presents an adaptive intention-driven variable impedance controller, where the online estimation of human motion intention is realized subject to physical interaction, stochastic distribution, and random disturbance. Specifically, human motion intention is estimated under a dual-channel structure and is represented as both the immediate desired position of the human limb and its predicted future position. A new variable impedance model is formulated from the estimated intention to regulate the dynamic interaction between the human and the robot. Such an impedance model is defined as the control objective achieved using the adaptive controller for wearable compliantly driven robots. The proposed formulation will be able to improve the estimation accuracy of human motion intention and will allow the robot to match the human’s action in a safe and efficient manner. The stability of the closed-loop system is rigorously proven from the stochastic perspective, and the experimental results on the compliantly driven exoskeleton robot are presented to validate the performance of the proposed controller.

(2023) Lower Limb Exoskeleton With Energy-Storing Mechanism for Spinal Cord Injury Rehabilitation, in IEEE Access, DOI: 10.1109/ACCESS.2023.3336308.
Keywords: Biomechanics; Body weight; brain-computer interface; Camshafts; Electroencephalography; Energy storage; Exoskeletons; Eye movements; gait analysis; Hip; Joints (anatomy); Knee; Legged locomotion; lower limb exoskeleton; medical robotics; Mobility; Potential energy; Prostheses; Quality of life; Reduction; Rehabilitation; robotic rehabilitation; Robots; Service robots; Spinal cord injuries; Spinal cord injury (SCI); Torque; Walking.
Abstract: Statistics from the National Office for Empowerment of Persons with Disabilities (NEP) indicate that Spinal Cord Injury (SCI) is a major cause of disability in the Thai population. Various rehabilitation methods are available to support SCI patients. Assistive robots, such as exoskeletons and prosthetics, are very useful for improving quality of life. Robotic exoskeletons have evolved as rehabilitation methods that can overcome some of the current health-related effects of SCI. In the current study, a lower-limb exoskeleton was developed to assist or rehabilitate a physically challenged person who has lost mobility owing to SCI. To overcome energy storage issues related to existing designs, the device uses a spring and camshaft system that is integrated with the robot structure to reduce the required energy by absorbing the body weight into spring potential energy and released by the cam design. Hence, the spring cam system significantly reduced torque on the joints, with approximately [Formula Omitted] reduction in the angle joint and [Formula Omitted] reduction in the knee joint. Control of the exoskeleton is carried out by analyzing brain signals (EEG) and eye movement signals (EOG), which are combined with the control system to perform daily activities, such as walking, turning, and standing. This exoskeleton boasts a maximum walking speed of 0.5 m/s and a remarkable two-hour full-load operation, making it a promising solution for enhancing the mobility and quality of life of individuals with SCI. The effectiveness of the developed exoskeleton in assisting individuals with mobility impairments was validated through comprehensive laboratory-level experimental analysis.;Statistics from the National Office for Empowerment of Persons with Disabilities (NEP) indicate that Spinal Cord Injury (SCI) is a major cause of disability in the Thai population. Various rehabilitation methods are available to support SCI patients. Assistive robots, such as exoskeletons and prosthetics, are very useful for improving quality of life. Robotic exoskeletons have evolved as rehabilitation methods that can overcome some of the current health-related effects of SCI. In the current study, a lower-limb exoskeleton was developed to assist or rehabilitate a physically challenged person who has lost mobility owing to SCI. To overcome energy storage issues related to existing designs, the device uses a spring and camshaft system that is integrated with the robot structure to reduce the required energy by absorbing the body weight into spring potential energy and released by the cam design. Hence, the spring cam system significantly reduced torque on the joints, with approximately $17-30\%$ reduction in the angle joint and $40-48\%$ reduction in the knee joint. Control of the exoskeleton is carried out by analyzing brain signals (EEG) and eye movement signals (EOG), which are combined with the control system to perform daily activities, such as walking, turning, and standing. This exoskeleton boasts a maximum walking speed of 0.5 m/s and a remarkable two-hour full-load operation, making it a promising solution for enhancing the mobility and quality of life of individuals with SCI. The effectiveness of the developed exoskeleton in assisting individuals with mobility impairments was validated through comprehensive laboratory-level experimental analysis.;Statistics from the National Office for Empowerment of Persons with Disabilities (NEP) indicate that Spinal Cord Injury (SCI) is a major cause of disability in the Thai population. Various rehabilitation methods are available to support SCI patients. Assistive robots, such as exoskeletons and prosthetics, are very useful for improving quality of life. Robotic exoskeletons have evolved as rehabilitation methods that can overcome some of the current health-related effects of SCI. In the current study, a lower-limb exoskeleton was developed to assist or rehabilitate a physically challenged person who has lost mobility owing to SCI. To overcome energy storage issues related to existing designs, the device uses a spring and camshaft system that is integrated with the robot structure to reduce the required energy by absorbing the body weight into spring potential energy and released by the cam design. Hence, the spring cam system significantly reduced torque on the joints, with approximately 17-30\% reduction in the angle joint and 40-48\% reduction in the knee joint. Control of the exoskeleton is carried out by analyzing brain signals (EEG) and eye movement signals (EOG), which are combined with the control system to perform daily activities, such as walking, turning, and standing. This exoskeleton boasts a maximum walking speed of 0.5 m/s and a remarkable two-hour full-load operation, making it a promising solution for enhancing the mobility and quality of life of individuals with SCI. The effectiveness of the developed exoskeleton in assisting individuals with mobility impairments was validated through comprehensive laboratory-level experimental analysis.

(2023) Predictive and Robust Robot Assistance for Sequential Manipulation, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2023.3320029.
Keywords: Assistive robots; Human-aware motion planning; Human-robot interaction; Motion planning; Multiple objective analysis; Optimal control; Optimization; optimization and optimal control; physical human-robot interaction; Prediction models; Robot kinematics; Robots; Robustness (mathematics); Task analysis.
Abstract: This letter presents a novel concept to support physically impaired humans in daily object manipulation tasks with a robot. Given a user’s manipulation sequence, we propose a predictive model that uniquely casts the user’s sequential behavior as well as a robot support intervention into a hierarchical multi-objective optimization problem. A major contribution is the prediction formulation, which allows to consider several different future paths concurrently. The second contribution is the encoding of a general notion of constancy constraints, which allows to consider dependencies between consecutive or far apart keyframes (in time or space) of a sequential task. We perform numerical studies, simulations and robot experiments to analyse and evaluate the proposed method in several table top tasks where a robot supports impaired users by predicting their posture and proactively re-arranging objects.

(2023) Bridging Human-Robot Co-Adaptation via Biofeedback for Continuous Myoelectric Control, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2023.3330053.
Keywords: Adaptability; Adaptation; Adaptation models; Adaptive control; Automatic control; Biofeedback; Biological system modeling; Completion time; Electrodes; EMG; Human-robot co-adaptation; Machine intelligence; Machine learning; Myoelectric control; Myoelectricity; Neural prostheses; Perturbation; Protocols; Recognition; Robot control; Robots; Robust control; Task analysis; Training; user intent recognition; Wrist.
Abstract: This letter proposes a novel human-robot co-adaptation framework for robust and accurate user intent recognition, specifically in the context of automatic control in assistance robots such as neural prosthetics and rehabilitation devices empowered by electrophysiological signals. Our goal is to incorporate user adaptability early in the training phase to facilitate both machine recognition and user adaptability, rather than relying solely on brute-force machine learning methods. The proposed framework is featured by applying biofeedback-based user adaptive behavior into model training, while the machine can adapt to those changes through online learning. Specifically, this study focuses on the recognition of two-degree-of-freedom simultaneous and continuous wrist movement intentions based on surface electromyogram (sEMG) array signals, and the performance is tested on twelve able-bodied subjects. The co-adaptive evaluation experiment demonstrates the robust control of this method by introducing sEMG electrode displacement as perturbations. Experimental results show that this method improves the completion time of centre-out tasks by 13% compared to conventional methods (Cohen’s d = 0.637), and debias 86% of the effect of electrode shift perturbations. This study provides insights into the potential for incorporating human adaptability into machine intelligence to improve user intent recognition and automatic robot control.

(2022) A Review of Human–Machine Cooperation in the Robotics Domain, in IEEE T. Hum.-Mach. Syst., DOI: 10.1109/THMS.2021.3131684.
Keywords: Robots; Robot sensing systems; Robot kinematics; Task analysis; Sensors; Decision making; Automation; Cooperative control; human–machine cooperation (HMC); human–machine integration; human–robot interaction; shared control.
Abstract: Artificial intelligence (AI) technology has greatly expanded human capabilities through perception, understanding, action, and learning. The future of AI depends on cooperation between humans and AI. In addition to a fully automated or manually controlled machine, a machine can work in tandem with a human with different levels of assistance and automation. Machines and humans cooperate in different ways. Three strategies for cooperation are described in this article, as well as the nesting relationships among different control methods and cooperation strategies. Based on human thinking and behavior, a hierarchical human–machine cooperation (HMC) framework is improved and extended to design safe, efficient, and attractive systems. We review the common methods of perception, decision-making, and execution in the HMC framework. Future applications and trends of HMC are also discussed.

(2022) Adaptive Cooperative Control for Human-Robot Load Manipulation, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2022.3158435.
Keywords: Adaptive control; Collaboration; Cooperative control; End effectors; Human-robot collaboration; Impedance; Jacobian matrices; Manipulator dynamics; multi-robot systems; Robot arms; Robot control; Robots; robust/adaptive control; Stability analysis; Stiffness; System dynamics; Task analysis; Uncertainty.
Abstract: In this letter, we propose a control strategy for human-robot cooperative manipulation under the ambiguous collaboration of a human agent. To cope with this uncertainty, an adaptive update law inferring the human contribution to the system dynamics from basic perception feedback through the human arm stiffness is used. Furthermore, the robustness and accuracy of the approach is enhanced by redundantly tracking the shared load references and its associated end-effector position references. To validate the control strategy, both theoretical Lyapunov stability analysis and experimental results –employing two robot manipulators with 6 degrees of freedom under external disturbances– are provided.;In this letter, we propose a control strategy forhuman-robot cooperative manipulation under the ambiguous collaboration of a human agent. To cope with this uncertainty, an adaptive update law inferring the human contribution to the system dynamics from basic perception feedback through the human arm stiffness is used. Furthermore, the robustness and accuracy of the approach is enhanced by redundantly tracking the shared load references and its associated end-effector position references. To validate the control strategy, both theoretical Lyapunov stability analysis and experimental results –employing two robot manipulators with 6 degrees of freedom under external disturbances– are provided.

(2022) Towards Safe Physical Human-Robot Interaction by Exploring the Rapid Stiffness Switching Feature of Discrete Variable Stiffness Actuation, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2022.3185366.
Keywords: Actuation; Actuators; Collision avoidance; Collision dynamics; Compliant Joints and Mechanisms; Gears; Human engineering; Physical Human-Robot Interaction; Robot dynamics; Robots; Safety; Safety in HRI; Springs; Stiffness; Switches; Switching; Torque.
Abstract: Safety holds the prime importance in direct physical human-robot interaction (pHRI) tasks. Robots should have the ability to handle unexpected collisions in unstructured environments. Collision avoidance based on exteroceptive sensors can work in these scenarios. However, it may not be sufficient, especially considering that the relative motion between robots and humans can be fast and hardly predictable. This highlights the importance of fast and reliable detection and reaction techniques for the collisions. Rapid switching to intrinsic complaint mode upon collisions is a promising solution for this requirement. Recently, we have proposed a new design of the discrete variable stiffness actuator (DVSA), which has the capability of instantaneously switching its stiffness between different predefined levels. We believe that this rapid stiffness switching feature can significantly improve safety during collisions. In this letter, we combined a software-based collision detection method with a hardware-based rapid stiffness switching technique. The proposed strategy has been implemented on a DVSA-based manipulator to evaluate its safety performance in the sudden dynamic collision and static near-singular clamping collision scenarios. The results clearly indicate that the proposed strategy can significantly mitigate the impact of unexpected collisions and improve safety during pHRI.

(2022) Low-Impedance Displacement Sensors for Intuitive Physical Human-Robot Interaction: Motion Guidance, Design, and Prototyping, in IEEE Trans. Robot., DOI: 10.1109/TRO.2021.3121610.
Keywords: Collaboration; Degrees of freedom; Displacement; Displacement sensor; Force; force balancing; Human motion; Impedance; intuitive interaction; low-impedance interaction; physical human–robot interaction; Prototyping; Robot dynamics; Robot kinematics; Robot sensing systems; Robots; Sensors.
Abstract: This article provides a general framework for the use of low-impedance displacement sensors mounted on the links of a serial robot to provide an intuitive physical human-robot interaction. A general formulation is developed to handle the motion guidance problem, i.e., the mapping of the measured motion of the sensors into the required robot joint motions to provide intuitive responsiveness. The formulation is general and can be applied to any architecture of serial robot with any number of displacement sensors each having an arbitrary number of degrees of freedom. Then, the design of a novel three-degree-of-freedom low-impedance displacement sensor is presented as a particularly effective instantiation of the general concept. Partial force balancing is used to reduce the required elastic return action, thereby ensuring the low impedance of the interaction. A prototype of a three-degree-of-freedom displacement sensor is then introduced. Two such sensors are mounted on the links of a custom-built five-degree-of-freedom robot in order to demonstrate the proposed approach. Experimental results are provided and comparisons with other collaborative robots are given. It is shown that the proposed sensors and motion guidance approach yield very intuitive low-impedance interaction involving very low interaction forces.

(2022) EMG-Based Variable Impedance Control With Passivity Guarantees for Collaborative Robotics, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2022.3149575.
Keywords: Collaboration; Compliance and Impedance Control; Computer Science; Electromyography; Force measurement; Human-Robot Collaboration; Impedance; Passivity; Physical Human-Robot Interaction; Robot control; Robot sensing systems; Robotics; Robots; Task analysis.
Abstract: In this paper, a new methodology is developed for safely changing the interaction dynamics of a collaborative robot. A strategy based on electromyography is proposed to distinguish operator forces from those resulting from interactions with the environment. This allows to obtain information about the operator intentions and include it into the robot control strategy for an enhanced physical human–robot interaction. The safety of the resulting variable impedance control is guaranteed by imposing the passivity of the interaction. Experimental validation shows a good performance of the proposed method and illustrates the advantages of such a strategy in cases where human, robot and environment interact with each other.

(2022) User-Adaptive Variable Damping Control Using Bayesian Optimization to Enhance Physical Human-Robot Interaction, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2022.3144511.
Keywords: Adaptive control; Assistive robotics; Bayesian analysis; Controllers; Damping; Force; Gaussian process; Human engineering; Human motion; Human performance; Impedance; impedance control; interaction control; Linear programming; Noise measurement; Optimization; Performance measurement; Physical human-robot interaction; Process parameters; Robot arms; Robots; Stability; Tuning.
Abstract: This letter presents a user-adaptive variable damping controller that enhances the overall performance of coupled human-robot systems in terms of stability, agility, user effort, and energy expenditure during physical human-robot interaction. The controller accounts for impedance properties of the human limbs and adaptively changes robotic damping from negative to positive values based on user’s intent of motion while minimizing energy of the coupled human-robot system. Bayesian optimization is used to evaluate an unknown objective function and optimize noisy performance, which builds on a Gaussian process to account for the uncertainty of human behaviors and noisy observations. To validate the effectiveness of the presented approach and evaluate its potential applications in real-world scenarios, we performed human experiments using a common robotic arm manipulator. Experimental results from five pilot subjects demonstrated that the controller does not require a long parameter tuning process. Compared to variable damping control without user-adaptive parameter changes, the presented adaptive control strategy could reduce ∼45% energy expenditure and achieve average performance improvement of ∼20% when several performance metrics of stability, agility, and user effort are considered together.

(2022) Macro-Mini Linear Actuator Using Electrorheological-Fluid Brake for Impedance Modulation in Physical Human-Robot Interaction, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2022.3145050.
Keywords: Actuator; Actuators; Antigravity; Body weight; Brakes; Concentric cylinders; electrorheological fluid; Electrorheological fluids; End effectors; Fluids; Force; Impedance; impedance control; Magnetorheological fluids; Mechanical impedance; Pneumatics; robotics; Robots; Rotors; Shafts; stroke rehabilitation; Support systems; Torque.
Abstract: Robots designed to interact physically with humans are typically characterized by low impedance and low output force, but in many circumstances, high force is needed. Integrated to a high-torque velocity source, antagonistic designs of electrorheological (ER)-fluid or magnetorheological (MR)-fluid clutches enabled a range of achievable impedances for rotational uses. This letter presents an alternative novel concept of linear actuator which uses a rotary ER-fluid brake to engage the highly backdrivable unit to the high force unit. The end effector driven by a mini motor is connected to the brake rotor. Since the brake rotation allows the relative translation between the endpoint and the high-force actuator, the mechanical impedance can be modulated by controlling the brake friction through the applied voltage. The ER-fluid brake using multiple concentric cylinders for high torque-to-inertia ratio was characterized experimentally. The macro-mini linear actuator with an intrinsic failsafe can be applied for active body weight support systems requiring antigravity high force.

(2022) Toward Safe Human-Robot Interaction: A Fast- Response Admittance Control Method for Series Elastic Actuator, in IEEE Trans. Autom. Sci. Eng., DOI: 10.1109/TASE.2021.3057883.
Keywords: Actuators; Admittance; Admittance control; Algorithms; Compensators; Control methods; delay compensation; Delay effects; Delays; Electrical impedance; Error compensation; Force; Human motion; Hydraulic systems; Impedance; manufacturing automation; Mechanical impedance; Modulus of elasticity; Optimization; Parameters; Robots; series elastic actuator (SEA); Stiffness; Time lag; Tracking errors.
Abstract: Series elastic actuator (SEA) is a promising compliance device due to its lower output mechanical impedance, and it is widely applied to ensure safe human–robot interaction. Although some efforts have been made to achieve accurate stiffness tracking, the time-delay issue in SEA control has still not been well investigated. However, the time delay can cause an inaccurate response and increase the risk of injury. To overcome this problem, this article proposes a fast-response admittance control method for SEAs. First, an admittance control scheme considering the external force estimation is developed for a hydraulic SEA. Then, a parallel adaptive time-series (ATS) (P-ATS) compensator is proposed and further adopted in the admittance control scheme to compensate for the time delay and tracking error. The P-ATS compensator is a modification of the ATS compensator, which is enhanced with a unique parallel mechanism. Such a mechanism can save more computational resources on locating better parameters the for P-ATS compensator, thus improving its performance. Moreover, the parameter setting is converted to an optimization task, which is solved by the whale swarm algorithm (WSA) to achieve higher accuracy. The newly located parameters are compared to the current parameters based on a proposed evaluation criterion, thus guaranteeing the quality of the updated parameters. All the above strategies are employed to improve the SEA admittance control performance. The results obtained from both simulation and real-world experiments validate that, compared to conventional methods, the proposed method achieves a better performance in SEA stiffness tracking with lower time delay and tracking error. Note to Practitioners —Accurate stiffness tracking of SEAs can achieve safe human–robot interaction. However, the time delays introduced by the imprecise movement and estimation of external force can lead to inaccurate actuator response that may limit the capacity of safety insurance. To overcome this issue, a fast-response admittance control method is proposed for SEAs by adopting a novel P-ATS compensator. Thus, the time delays and errors from both load movement and external force estimation can be adaptively compensated. Several strategies have been adopted to enhance the compensator for parameter determination to achieve better performance. The proposed method requires no additional previous information about the system except load mass and spring stiffness, which makes it easy to implement for different types of SEAs. Experimental results show that the proposed method can achieve faster and more accurate stiffness tracking under different conditions. Future work aims to address the control problem under random disturbances and apply the proposed method to human–robot collaboration tasks to further test its performance.

(2022) Dynamic Primitives Limit Human Force Regulation During Motion, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2022.3141778.
Keywords: Actuators; Bars; compliance and impedance control; Computer Science; Errors; Exoskeletons; Feedback; Force; Force control; Human motion; Hybrid control; Mechanical impedance; Physical human-robot interaction; Prostheses; Rehabilitation robots; Robot sensing systems; Robotics; Robots; Task analysis; Visualization.
Abstract: Humans excel at physical interaction despite long feedback delays and low-bandwidth actuators. Yet little is known about how humans manage physical interaction. A quantitative understanding of how they do is critical for designing machines that can safely and effectively interact with humans, e.g. amputation prostheses, assistive exoskeletons, therapeutic rehabilitation robots, and physical human-robot collaboration. To facilitate applications, this understanding should be in the form of a simple mathematical model that not only describes humans’ capabilities but also their limitations. In robotics, hybrid control allows simultaneous, independent control of both motion and force and it is often assumed that humans can modulate force independent of motion as well. This paper experimentally tested that assumption. Participants were asked to apply a constant 5N force on a robot manipulandum that moved along an elliptical path. After initial improvement, force errors quickly plateaued, despite practice and visual feedback. Within-trial analyses revealed that force errors varied with position on the ellipse, rejecting the hypothesis that humans have independent control of force and motion. The findings are consistent with a feed-forward motion command composed of two primitive oscillations acting through mechanical impedance to evoke force.;Humans excel at physical interaction despite long feedback delays and low-bandwidth actuators. Yet little is known about how humans manage physical interaction. A quantitative understanding of how they do is critical for designing machines that can safely and effectively interact with humans, e.g. amputation prostheses, assistive exoskeletons, therapeutic rehabilitation robots, and physical human-robot collaboration. To facilitate applications, this understanding should be in the form of a simple mathematical model that not only describes humans’ capabilities but also their limitations. In robotics, hybrid control allows simultaneous, independent control of both motion and force and it is often assumed that humans can modulate force independent of motion as well. This letter experimentally tested that assumption. Participants were asked to apply a constant 5 N force on a robot manipulandum that moved along an elliptical path. After initial improvement, force errors quickly plateaued, despite practice and visual feedback. Within-trial analyses revealed that force errors varied with position on the ellipse, rejecting the hypothesis that humans have independent control of force and motion. The findings are consistent with a feed-forward motion command composed of two primitive oscillations acting through mechanical impedance to evoke force.

(2022) Adaptive-Constrained Impedance Control for Human-Robot Co-Transportation, in IEEE T. Cybern., DOI: 10.1109/TCYB.2021.3107357.
Keywords: Actuators; Adaptive algorithms; Adaptive control; Collaboration; Constraints; Control algorithms; Control theory; Controllers; End effectors; Error constraint; Force feedback; Human motion; Human-robot interaction; human–robot co-transportation; Impedance; Initial conditions; input constraint; Interaction models; Neural networks; neural networks (NNs); Nonlinear control; Position measurement; Robot control; Robot dynamics; Robot sensing systems; Robots; Task space; Transportation; vision and force sensing.
Abstract: Human-robot co-transportation allows for a human and a robot to perform an object transportation task cooperatively on a shared environment. This range of applications raises a great number of theoretical and practical challenges arising mainly from the unknown human-robot interaction model as well as from the difficulty of accurately model the robot dynamics. In this article, an adaptive impedance controller for human-robot co-transportation is put forward in task space. Vision and force sensing are employed to obtain the human hand position, and to measure the interaction force between the human and the robot. Using the latest developments in nonlinear control theory, we propose a robot end-effector controller to track the motion of the human partner under actuators’ input constraints, unknown initial conditions, and unknown robot dynamics. The proposed adaptive impedance control algorithm offers a safe interaction between the human and the robot and achieves a smooth control behavior along the different phases of the co-transportation task. Simulations and experiments are conducted to illustrate the performance of the proposed techniques in a co-transportation task.

(2022) Assimilation Control of a Robotic Exoskeleton for Physical Human-Robot Interaction, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2022.3144537.
Keywords: Adaptive control; Assimilation; assimilation control; Collaboration; Control methods; Control stability; Control systems design; Dynamics; Exoskeletons; Force; Human engineering; Human motion; Human performance; intention recognition; Obstacle avoidance; Physical human-robot interaction; Robot control; Robots; safety in HRI; Task analysis; Trajectory.
Abstract: The ability of human operators in estimating the partners’ motion intention and utilizing it for collaboration brings valuable enlightenment to human-robot systems. Motivated by these observations, this letter introduces an assimilation control method that reshapes the physical interaction trajectory in the interaction task, which enables the exoskeleton robot to estimate the subject’s virtual target from the interaction force and adapt its own behavior. Under the assumption that the virtual target is determined by the control gains, the stability of the human-robot system is guaranteed, and the proposed scheme realizes continuous interaction behaviors from cooperation to competition. Then an adaptive controller is designed to enable the robot to directly deal with uncertain dynamics and joint space constraints. The experiment verifies how the assimilation control method assists the subjects in a collaborative execution or gradually competes with them to avoid collisions. Compared with related literature, our approach is able to realize safe manipulation (e.g., obstacle avoidance) and broader interaction behaviors by reshaping the interactive trajectory using force feedback only, without continuous manual guidance as in many existing methods.

(2022) Spatial Iterative Learning Control for Robotic Path Learning, in IEEE T. Cybern., DOI: 10.1109/TCYB.2021.3138992.
Keywords: Control methods; Force; Iterative learning control; Learning; Learning law; path learning; Robot control; Robot kinematics; Robot sensing systems; Robots; spatial iterative learning control (sILC); surface exploration; Task analysis; teaching by demonstration; Trajectory; Unknown environments.
Abstract: A spatial iterative learning control (sILC) method is proposed for a robot to learn a desired path in an unknown environment. When interacting with the environment, the robot initially starts with a predefined trajectory so an interaction force is generated. By assuming that the environment is subjected to fixed spatial constraints, a learning law is proposed to update the robot’s reference trajectory so that a desired interaction force is achieved. Different from existing iterative learning control methods in the literature, this method does not require repeating the interaction with the environment in time, which relaxes the assumption of the environment and thus addresses the limits of the existing methods. With the rigorous convergence analysis, simulation and experimental results in two applications of surface exploration and teaching by demonstration illustrate the significance and feasibility of the proposed method.

(2022) Fuzzy Enhanced Adaptive Admittance Control of a Wearable Walking Exoskeleton With Step Trajectory Shaping, in IEEE Trans. Fuzzy Syst., DOI: 10.1109/TFUZZ.2022.3162700.
Keywords: Adaptive control; Adaptive fuzzy control; admittance control; Control systems design; Controllers; Disturbance observers; Dynamic models; Electrical impedance; Exoskeletons; Force; Fuzzy control; Gait; Hip; Human motion; Legged locomotion; Liapunov functions; Perturbation; Robot kinematics; Robots; step trajectory shaping; Trajectory; Walking; walking exoskeleton.
Abstract: The generation of motor adaptation in response to mechanical perturbation during human walking is seldom considered in an exoskeleton system. Reshaping step trajectory over consecutive gait cycles for a walking exoskeleton is investigated in this article. Step adjustment of a walking exoskeleton can adapt to human walking intention by shaping step trajectory. This work develops an admittance adaptive fuzzy control strategy for a walking exoskeleton robot to provide assistance for human lower limb movement. Considering human walking intention and utilizing an admittance model, it shapes a reference trajectory to ensure that the walking exoskeleton follows it according to the human–robot force produced by its wearer. Considering a nonlinear and dynamic model with uncertainties, this work designs an integral-type Lyapunov function controller to track a reference trajectory. A disturbance observer is integrated into the controller design to compensate for uncertain disturbance in order to achieve an effective tracking performance. Finally, this work conducts experiments on two healthy subjects with the proposed method on a walking exoskeleton to validate its effectiveness. The results show that it can be applied to walking exoskeletons to enhance human mobility.

(2022) Learning-Based Approach for a Soft Assistive Robotic Arm to Achieve Simultaneous Position and Force Control, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2022.3185786.
Keywords: Actuators; and learning for soft robots; Bathing; control; Control methods; First principles; Force; Learning; Manipulators; model learning for control; Modeling; Nonlinearity; optimization and optimal control; People with disabilities; Probabilistic logic; Probabilistic models; Robot arms; Robotics; Robots; Soft robotics; Task analysis; Uncertainty.
Abstract: Soft robotics have demonstrated great advantages in assisting elderly/disabled people during daily tasks, owing to their highly dexterous motions and safe human-robot interactions. However, simultaneously controlling the position and force of soft robots is still a challenging task due to soft actuators’ nonlinearity, system uncertainty, and high-dimensional control space. Classical control methods are usually based on first-principle/analytical models that are difficult to derive for soft robots without making significant simplifications. To overcome such control challenges, the central concept of this work is to introduce a learning-based data-driven approach. The approach employs a probabilistic model to explicitly capture system nonlinearity and uncertainty. Besides, nonparametric local learning methods are investigated to deal with redundant high-dimensional control space. The approach is applied in a soft robotic arm interacting with a manikin to simulate the bathing task. Experimental results demonstrate that the soft robotic arm could be well controlled to track the desired position and force simultaneously (maximum error of position and force is 6 mm and 0.037 N). Meanwhile, our method outperforms another typical data-driven approach (maximum error of position and force is 10 mm and 0.058 N). The results indicate that our approach is helpful for soft robots because of the physical interactions needed in assistive tasks.

(2022) A Control Framework for Adaptation of Training Task and Robotic Assistance for Promoting Motor Learning With an Upper Limb Rehabilitation Robot, in IEEE transactions on systems, man, and cybernetics. Systems, DOI: 10.1109/TSMC.2022.3163916.
Keywords: Adaptive algorithms; Adaptive control; Assist as needed (AAN); Assistive robots; Feature extraction; Feedback control; Feedforward control; Learning; motor learning; nonlinear adaptive control; Oscillators; Performance evaluation; Radial basis function; Real-time systems; Rehabilitation; Rehabilitation robotics; Rehabilitation robots; Robot control; Robotics; Robots; Trajectory; Trajectory optimization.
Abstract: Robot-assisted rehabilitation has been a promising solution to improve motor learning of neurologically impaired patients. State-of-the-art control strategies are typically limited to the ignorance of heterogeneous motor capabilities of poststroke patients and therefore intervene suboptimally. In this article, we propose a control framework for robot-assisted motor learning, emphasizing the detection of human intention, generation of reference trajectories, and modification of robotic assistance. A real-time trajectory generation algorithm is presented to extract the high-level features in active arm movements using an adaptive frequency oscillator (AFO) and then integrate the movement rhythm with the minimum-jerk principle to generate an optimal reference trajectory, which synchronizes with the motion intention in the patient as well as the motion pattern in healthy humans. In addition, a subject-adaptive assistance modification algorithm is presented to model the patient’s residual motor capabilities employing spatially dependent radial basis function (RBF) networks and then combining the RBF-based feedforward controller with the impedance feedback controller to provide only necessary assistance while simultaneously regulating the maximum-tolerated error during trajectory tracking tasks. We conduct simulation and experimental studies based on an upper limb rehabilitation robot to evaluate the overall performance of the motor-learning framework. A series of results showed that the difficulty level of reference trajectories was modulated to meet the requirements of subjects’ intended motion, furthermore, the robotic assistance was compliantly optimized in response to the changing performance of subjects’ motor abilities, highlighting the potential of adopting our framework into clinical application to promote patient-led motor learning.

(2022) Iterative Learning-Based Robotic Controller With Prescribed Human-Robot Interaction Force, in IEEE Trans. Autom. Sci. Eng., DOI: 10.1109/TASE.2021.3119400.
Keywords: Automation; Control methods; Controllers; Exoskeletons; Force; Force tracking; Human motion; Human-robot interaction; iterative learning; Iterative learning control; Learning; Optimization; Performance indices; Rehabilitation robots; Robot control; robotic controller; Robots; Trajectories; Uncertainty.
Abstract: In this article, an iterative-learning-based robotic controller is developed, which aims at providing a prescribed assistance or resistance force to the human user. In the proposed controller, the characteristic parameter of the human upper limb movement is first learned by the robot using the measurable interaction force, a recursive least square (RLS)-based estimator, and the Adam optimization method. Then, the desired trajectory of the robot can be obtained, tracking which the robot can supply the human’s upper limb with a prescribed interaction force. Using this controller, the robot automatically adjusts its reference trajectory to embrace the differences between different human users with diverse degrees of upper limb movement characteristics. By designing a performance index in the form of interaction force integral, potential adverse effects caused by the time-related uncertainty during the learning process can be addressed. The experimental results demonstrate the effectiveness of the proposed method in supplying the prescribed interaction force to the human user. Note to Practitioners—This article concentrates on developing a novel control technique to make the robot supply a prescribed interaction force to the human user in the presence of time-related uncertainties. The proposed control method is applicable to various scenarios of the human–robot interaction, e.g., it can be used for rehabilitation robots to provide assistive or resistive force to stroke patients or for exoskeleton robots to provide assistive force to human users for completing heavy-load tasks. Moreover, the desired interaction force can be tailored for different human users according to their needs and different task objectives. Consequently, the proposed controller can serve diverse users and has a promising perspective in automation.

(2022) Impedance Learning-Based Adaptive Control for Human-Robot Interaction, in IEEE Trans. Control Syst. Technol., DOI: 10.1109/TCST.2021.3107483.
Keywords: Adaptation; Adaptive control; Autonomous impedance variation; Controllers; Dynamics; Feedback control; Force; Impedance; Learning; Moving targets; nonlinear adaptive control; physical human–robot interaction (pHRI); Robot control; Robot sensing systems; robot stability; Robots; Stability analysis; Task analysis; Uncertainty; uniform ultimate boundedness (UUB).
Abstract: In this article, a new learning-based time-varying impedance controller is proposed and tested to facilitate an autonomous physical human-robot interaction (pHRI). Novel adaptation laws are formulated for online adjustment of robot impedance based on human behavior. Two other sets of update rules are defined for intelligent coping with the robot’s structured and unstructured uncertainties. These rules ensure stability via Lyapunov’s theorem and provide uniform ultimate boundedness (UUB) of the closed-loop system’s response, without a need for HRI force/torque measurement. Accordingly, the convergence of response signals, including errors in tracking, online impedance learning, robot parameter adaptation, and controller gain variation, is proven to operate in a bounded region (compact set) in the presence of robot and human uncertainties and bounded disturbances. The performance of the developed intelligent impedance-varying control strategy is investigated through comprehensive experimental studies in a repetitive following task with a moving target.

(2022) Field-Based Human-Centred Control on SO(3) for Assist-as-Needed Robotic Rehabilitation, in IEEE transactions on medical robotics and bionics, DOI: 10.1109/TMRB.2022.3194372.
Keywords: Aerospace electronics; Algorithms; Angular velocity; angular velocity field; Attitude control; Biomimetics; contour following; Control methods; Control systems design; Controllers; Deviation; Gait; Human motion; Posture; Rehabilitation; Rehabilitation robotics; Robots; Torque; Training; Trajectory; Velocity distribution.
Abstract: Assist-as-needed (AAN) control can effectively improve the participation of the patient. However, the current research on attitude control in a manifold and also a special orthogonal group, SO(3), adapted to three-dimensional (3D) gait training and three-degree-of-freedom joint training is insufficient. The controller drives the patients to follow a timed trajectory and have the disadvantage of imposing a defined timing of movement. The deviation between patient and robot is neglected and the actual movement of human is not considered. Therefore, this study proposes a human-centred control method on SO(3) for AAN robotic rehabilitation. First, a feedback-stabilized closest attitude tracking algorithm is proposed to establish the angular velocity field and moment field on SO(3). And the framework of the field-based double-loop control is realised. The posture information of the human limbs is fed back to the controller through the sensing system, and the motion state of the human body is sensed in real-time, which is used as the basis for adjusting the torque to realise human-centred control. The experimental results show that different levels of adjustment torque are generated based on attitude deviations between the human and robot to verify the effectiveness of the designed controller.

(2022) Impedance Variation and Learning Strategies in Human-Robot Interaction, in IEEE T. Cybern., DOI: 10.1109/TCYB.2020.3043798.
Keywords: Impedance; Robots; Task analysis; Robot kinematics; Force; Damping; End effectors; Human–robot interaction (HRI); impedance and admittance models; impedance control; impedance learning; impedance variation; robot learning; robot stability.
Abstract: In this survey, various concepts and methodologies developed over the past two decades for varying and learning the impedance or admittance of robotic systems that physically interact with humans are explored. For this purpose, the assumptions and mathematical formulations for the online adjustment of impedance models and controllers for physical human-robot interaction (HRI) are categorized and compared. In this systematic review, studies on: 1) variation and 2) learning of appropriate impedance elements are taken into account. These strategies are classified and described in terms of their objectives, points of view (approaches), and signal requirements (including position, HRI force, and electromyography activity). Different methods involving linear/nonlinear analyses (e.g., optimal control design and nonlinear Lyapunov-based stability guarantee) and the Gaussian approximation algorithms (e.g., Gaussian mixture model-based and dynamic movement primitives-based strategies) are reviewed. Current challenges and research trends in physical HRI are finally discussed.

(2022) Impedance Control of a Wrist Rehabilitation Robot Based on Autodidact Stiffness Learning, in IEEE transactions on medical robotics and bionics, DOI: 10.1109/TMRB.2022.3194528.
Keywords: Actuators; Anatomical stiffness prediction; biomimetic muscle actuators (BMA); Biomimetics; Controllers; Dynamic control; End effectors; Human motion; Impedance; impedance control; Joints (anatomy); Koopman operator; Medical treatment; Muscles; non-linear control; Parallel robots; Rehabilitation robots; Robots; Stiffness; System identification; Wrist; wrist rehabilitation robot.
Abstract: Dynamic control of an intrinsically compliant robot is paramount to ensuring safe and synergistic assistance to the patient. This paper presents an impedance controller for the rehabilitation of stroke patients with compromised wrist motor functions. The control design employs a Koopman operator-based autodidactic system identification model to predict the anatomical stiffness of the wrist joint during its various degrees of rotational motion. The proposed impedance controller, perceiving the level of the subjects’ participation from their joint stiffness, can modify the applied force. The end-effector robot has a parallel structure that uses four biomimetic muscle actuators as parallel links between the end-effector and the base platform. The controller performance is corroborated by testing the end-effector robot with three healthy subjects.

(2022) Model-Based Actor−Critic Learning of Robotic Impedance Control in Complex Interactive Environment, in IEEE Trans. Ind. Electron., DOI: 10.1109/TIE.2021.3134082.
Keywords: Impedance; Robots; Force; Indexes; Task analysis; Safety; Service robots; Actor−critic learning; human−robot interaction; impedance control; robot machining.
Abstract: In complex robot applications, such as human−robot interaction and robot machining, robots should interact with an unknown environment. To learn the interactive skill, a model-based actor−critic learning algorithm and a safety-learning strategy are proposed in this article to find the optimal impedance control, in which the learning process is safe and fully automatic and does not know the system parameter. In the learning algorithm, a critic is defined as a quadratic form of the system states and the external force. A modified deterministic policy gradient algorithm is presented to improve the learning efficiency. The proposed approach utilizes a model-based constraint and a highly efficient learning algorithm. In the safety-learning strategy, the robot is trained under a constant force, and the learned impedance control can transfer to different interaction situations by choosing the suitable impedance index. The effectiveness of the learning algorithm and the performance of the learned impedance control are validated in a UR5 robot. The robot can perform human−robot interaction and robot machining tasks after the training process with 100 s training time.

(2022) A Proactive Controller for Human-Driven Robots Based on Force/Motion Observer Mechanisms, in IEEE transactions on systems, man, and cybernetics. Systems, DOI: 10.1109/TSMC.2022.3143892.
Keywords: Controllers; Dynamics; Force; Force control; Human motion; human–robot interaction (HRI); Impedance; motion control; Robot dynamics; Robot sensing systems; Robots; Trajectory; Uncertainty.
Abstract: This article investigates human-driven robots via physical interaction, which is enhanced by integrating the human partner’s motion intention. A human motor control model is employed to estimate the human partner’s motion intention. A system observer is developed to estimate the human’s control input in this model, so that force sensing is not required. A robot controller is developed to incorporate the estimated human’s motion intention, which makes the robot proactively follow the human partner’s movements. Simulations and experiments on a physical robot are carried out to demonstrate the properties of our proposed controller.

(2021) Vector Field Control Methods for Discretely Variable Passive Robotic Devices, in IEEE Trans. Robot., DOI: 10.1109/TRO.2020.3031255.
Keywords: Brakes; Continuously variable; Control methods; Degrees of freedom; Devices; Fields (mathematics); Force; Haptics; human-robot interaction; Hydraulic systems; Hydraulic transmissions; Manifolds; Manifolds (mathematics); Operators (mathematics); rehabilitation robotics; robot control; Robots; Steering; Switches; Task analysis; Velocity distribution.
Abstract: Passive transmission-based robotic devices are capable of providing motion guidance while ensuring user safety and engagement. To circumvent some of the drawbacks associated with steering continuously variable transmissions based on rolling contacts, we are exploring a class of discretely variable devices, based on brakes and hydrostatic transmissions. Previously available control methods for discretely variable devices were built on velocity fields and only developed to stabilize a one-dimensional (1-D) target manifold. For [Formula Omitted]-DOF devices, methods to stabilize target manifolds of dimensions 1 to [Formula Omitted]1 are of interest. In this article, we contribute constraint field methods that stabilize n [Formula Omitted]1 dimensional target manifolds while leaving the orthogonal subspace free to the control of the operator. We also contribute force-modulated single degree of freedom (DOF) velocity fields, which add between 1 and [Formula Omitted]2 virtual DOFs to the motion of devices whose physical constraints leave one DOF. Control performance is demonstrated in simulation for 3-DOF devices capable of imposing 1-D or 2-D constraints and in experiment for 2-DOF devices imposing 1-D constraints. Our experimental apparatus features digital hydraulic transmissions that are easily configured for an n -dimensional space and capable of imposing constraints of any dimension, thus motivating the contributed methods.;Passive transmission-based robotic devices are capable of providing motion guidance while ensuring user safety and engagement. To circumvent some of the drawbacks associated with steering continuously variable transmissions based on rolling contacts, we are exploring a class of discretely variable devices, based on brakes and hydrostatic transmissions. Previously available control methods for discretely variable devices were built on velocity fields and only developed to stabilize a one-dimensional (1-D) target manifold. For n-DOF devices, methods to stabilize target manifolds of dimensions 1 to n-1 are of interest. In this article, we contribute constraint field methods that stabilize n -1 dimensional target manifolds while leaving the orthogonal subspace free to the control of the operator. We also contribute force-modulated single degree of freedom (DOF) velocity fields, which add between 1 and n-2 virtual DOFs to the motion of devices whose physical constraints leave one DOF. Control performance is demonstrated in simulation for 3-DOF devices capable of imposing 1-D or 2-D constraints and in experiment for 2-DOF devices imposing 1-D constraints. Our experimental apparatus features digital hydraulic transmissions that are easily configured for an n -dimensional space and capable of imposing constraints of any dimension, thus motivating the contributed methods.

(2021) Applying Hip Stiffness With an Exoskeleton to Compensate Gait Kinematics, in IEEE Trans. Neural Syst. Rehabil. Eng., DOI: 10.1109/TNSRE.2021.3132621.
Keywords: Adaptation; Aging; Biomechanical Phenomena; Central nervous system; Exoskeleton; Exoskeleton Device; Exoskeletons; Gait; gait kinematics; Hip; hip exoskeleton robot; Humans; Kinematics; Legged locomotion; locomotor rehabilitation; Lower Extremity; Lower-limb exoskeleton robot; Motor task performance; Neurological diseases; Recovery; Rehabilitation; Robots; Stiffness; Technology; Therapeutic applications; Thigh; Torque; Walking.
Abstract: Neurological disorders and aging induce impaired gait kinematics. Despite recent advances, effective methods using lower-limb exoskeleton robots to restore gait kinematics are as yet limited. In this study, applying virtual stiffness using a hip exoskeleton was investigated as a possible method to guide users to change their gait kinematics. With a view to applications in locomotor rehabilitation, either to provide assistance or promote recovery, this study assessed whether imposed stiffness induced changes in the gait pattern during walking; and whether any changes persisted upon removal of the intervention, which would indicate changes in central neuro-motor control. Both positive and negative stiffness induced immediate and persistent changes of gait kinematics. However, the results showed little behavioral evidence of persistent changes in neuro-motor control, not even short-lived aftereffects. In addition, stride duration was little affected, suggesting that at least two dissociable layers exist in the neuro-motor control of human walking. The lack of neuro-motor adaptation suggests that, within broad limits, the central nervous system is surprisingly indifferent to the details of lower limb kinematics. The lack of neuro-motor adaptation also suggests that alternative methods may be required to implement a therapeutic technology to promote recovery. However, the immediate, significant, and reproducible changes in kinematics suggest that applying hip stiffness with an exoskeleton may be an effective assistive technology for compensation.

(2021) Motion Planning for Human-Robot Collaboration Using an Objective-Switching Strategy, in IEEE T. Hum.-Mach. Syst., DOI: 10.1109/THMS.2021.3112953.
Keywords: Adaptive control; Cognitive human–robot interaction; Collaboration; collaborative robot; Collision avoidance; Feasibility; Human motion; Human-robot interaction; learning and adaptive systems; Motion planning; Occupational safety; Robot dynamics; Strategy; Switching; Trajectory.
Abstract: Motion planning of collaborative robots is often required to simultaneously satisfy the contradictory objectives of reliably avoiding human workers and safely approaching them. In this article, we propose a new strategy that adaptively selects one of two objective functions based on the current operational region of the robot; the objective function for avoiding the worker with a distance larger than the safe distance, and the objective function for approaching the worker with a speed limitation for human safety. This strategy improves the worker safety while limiting the negative impact of the speed limit on time efficiency. The chattering related to the switching between the two objectives is solved using a continuous approximation of the switching based on uncertainties of the predicted worker’s motion. We implement the proposed motion planning with the objective-switching strategy into a collaborative assembly system to deal with the worker moving with high irregularities. We experimentally evaluate the effectiveness of the proposed motion planning from the safety and efficiency points of view.

(2021) Adaptive Human Force Scaling via Admittance Control for Physical Human-Robot Interaction, in IEEE Trans. Haptics, DOI: 10.1109/TOH.2021.3071626.
Keywords: Admittance; Force; Human factors; Collaboration; Admittance control; Human-robot interaction; Physical human-robot interaction; collaborative manipulation; adaptive force amplification; admittance control; human intention; Fitts’ task.
Abstract: The goal of this article is to design an admittance controller for a robot to adaptively change its contribution to a collaborative manipulation task executed with a human partner to improve the task performance. This has been achieved by adaptive scaling of human force based on her/his movement intention while paying attention to the requirements of different task phases. In our approach, movement intentions of human are estimated from measured human force and velocity of manipulated object, and converted to a quantitative value using a fuzzy logic scheme. This value is then utilized as a variable gain in an admittance controller to adaptively adjust the contribution of robot to the task without changing the admittance time constant. We demonstrate the benefits of the proposed approach by a pHRI experiment utilizing Fitts’ reaching movement task. The results of the experiment show that there is a) an optimum admittance time constant maximizing the human force amplification and b) a desirable admittance gain profile which leads to a more effective co-manipulation in terms of overall task performance.

(2021) A Framework for Autonomous Impedance Regulation of Robots Based on Imitation Learning and Optimal Control, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2020.3033260.
Keywords: Cartesian coordinates; Ellipsoids; Geometry; Imitation learning; Impedance; Linear quadratic regulator; Optimal control; optimization and optimal control; physical human-robot interaction; Probabilistic models; Robots; Stability analysis; Stiffness; Task analysis; Tracking errors; Trajectory.
Abstract: In this work, we propose a framework to address the autonomous impedance regulation problem of robots in a class of constrained manipulation tasks. In this framework, a human arm endpoint stiffness model is used to extract the task stiffness geometry along the constrained trajectory, which is then encoded offline and reproduced online by a Gaussian Mixture Model (GMM) and the Gaussian Mixture Regression (GMR), respectively. Furthermore, the full Cartesian impedance of the robot is formulated through an optimal control problem, i.e., the Linear-Quadratic Regulator (LQR), in which the task stiffness geometry (extracted from human demonstrations) is considered as the time-varying weighting matrix Q. The optimal impedance is eventually realised by the robot through a task geometry consistent Cartesian impedance controller. A tank-based passivity observer is implemented to give evidence on the stability of the system during online impedance variations. To evaluate the performance of the framework, a comparative experiment with three different impedance settings (i.e., the proposed framework, the framework without LQR and the framework without GMM/GMR) for Franka Emika Panda to perform a door opening task was conducted. The results reveal that our framework outperforms the other two, in terms of tracking error and the interaction forces.

(2021) An Investigation of a Balanced Hybrid Active-Passive Actuator for Physical Human-Robot Interaction, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2021.3064497.
Keywords: Actuation; Actuators; Brakes; D C motors; Electric motors; Force; Human engineering; Human-robot interaction; Impedance; Industrial robots; Robots; Safety; Torque; Trajectory optimization.
Abstract: Cooperative robots or “cobots” promise to allow humans and robots to work together more closely while maintaining safety. However, to date the capabilities of cobots are greatly diminished compared to industrial robots in terms of the force and power they are able to safely produce. This is in part due to the actuation choices of cobots. Low impedance robotic actuators aim to solve this problem by attempting to provide an actuator with a combination of low output impedance and a large bandwidth of force control. In short the ideal actuator has a large dynamic range. Existing actuators success and performance has been limited. We propose a high force and high power balanced hybrid active-passive actuator which aims to increase the actuation capability of low impedance actuators and to safely enable high performance larger force and workspace robots. Our balanced hybrid actuator does so, by combining and controlling a series elastic actuator, a small DC motor, and a particle brake in parallel. The actuator provides low and high frequency power producing active torques, along with power absorbing passive torques. Control challenges and advantages of hybrid actuators are discussed and overcome through the use of trajectory optimization, and the safety of the new actuator is evaluated.

(2021) Optimized Interaction Control for Robot Manipulator Interacting With Flexible Environment, in IEEE/ASME transactions on mechatronics, DOI: 10.1109/TMECH.2020.3047919.
Keywords: Cost function; Error analysis; Flexible environment; Force; Human-robot interaction; improved Q-learning method; Machine learning; Manipulator dynamics; optimized interaction control; Parameters; Q-learning; Robot arms; Robot control; robot-environment interaction; Robots; Surface impedance; Task analysis; Tracking errors; Trajectory; Trajectory measurement; Unknown environments.
Abstract: In this article, a novel interactioncontrol is presented to resolve the optimized robot-environment interaction control problems subject to flexible environment with unknown dynamics parameters. A cost function measuring the trajectory tracking error as well as the noninertial interaction force is defined. A complete state-space equation considering the robot desired trajectory, object dynamics and position parameters is also presented to address the optimized robot-environment interaction control problem. The improved Q-learning method is developed as the fundamental of the proposed control to deal with the challenges brought by the unknown environment dynamics and the reference position of the robot desired trajectory. Simulation and experimental studies verify the validity of the presented method.

(2021) Bayesian Estimation of Human Impedance and Motion Intention for Human-Robot Collaboration, in IEEE T. Cybern., DOI: 10.1109/TCYB.2019.2940276.
Keywords: Adaptive control; Adaptive impedance control; Bayes methods; Bayesian analysis; Bayesian estimation; Collaboration; Dynamics; Estimation; Force; Gaussian distribution; human impedance; Human motion; human motion intention estimation; Impedance; Neural networks; neural networks (NNs); Normal distribution; Robot dynamics; Robots; Stiffness; Task analysis; Tracking.
Abstract: This article proposes a Bayesian method to acquire the estimation of human impedance and motion intention in a human-robot collaborative task. Combining with the prior knowledge of human stiffness, estimated stiffness obeying Gaussian distribution is obtained by Bayesian estimation, and human motion intention can be also estimated. An adaptive impedance control strategy is employed to track a target impedance model and neural networks are used to compensate for uncertainties in robotic dynamics. Comparative simulation results are carried out to verify the effectiveness of estimation method and emphasize the advantages of the proposed control strategy. The experiment, performed on Baxter robot platform, illustrates a good system performance.

(2021) Predictive Exoskeleton Control for Arm-Motion Augmentation Based on Probabilistic Movement Primitives Combined With a Flow Controller, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2021.3068892.
Keywords: Assistive devices; Control methods; Controllers; Exoskeletons; Human-robot interaction; Movement; Physical human-robot interaction; physically assistive devices; Predictive control; Prosthetics; prosthetics and exoskeletons; Velocity distribution.
Abstract: There are many work-related repetitive tasks where the application of exoskeletons could significantly reduce the physical effort by assisting the user in moving the arms towards the desired location in space. To make such control more user acceptable, the controller should be able to predict the motion of the user and act accordingly. This letter presents an exoskeleton control method that utilizes probabilistic movement primitives to generate predictions of user movements in real-time. These predictions are used in a flow controller, which represents a novel velocity-field-based exoskeleton control approach to provide assistance to the user in a predictive way. We evaluated our approach with a haptic robot, where a group of twelve participants had to perform movements towards different target locations in the frontal plane. We tested whether we could generalize the predictions for new and unknown target locations whilst providing assistance to the user without changing their kinematic parameters. The evaluation showed that we could accurately predict user movement intentions while at the same time significantly decrease the overall physical effort exerted by the participants to achieve the task.

(2021) Lower-Limb Exoskeleton With Variable-Structure Series Elastic Actuators: Phase-Synchronized Force Control for Gait Asymmetry Correction, in IEEE Trans. Robot., DOI: 10.1109/TRO.2020.3034017.
Keywords: Actuators; Algorithms; Asymmetry; Cables; Control methods; Exoskeleton; Exoskeletons; Force; force control; Gait; Knee; Legged locomotion; medical robotics; rehabilitation robotics; Riccati equation; Robot control; Robots; Springs; Stiffness; Switches; Torque; Tracking control.
Abstract: Series elastic actuators (SEAs) can provide accurate force control and backdrivability in physical human–robot interaction. Control of SEA-generated forces or torques makes allowance for the user’s own volitional control and allows implementing a wide variety of assistive strategies. A novel force control method for a SEA-driven lower-limb assistive exoskeleton is presented. The device features variable-structure SEAs coupled via Bowden cables. The actuator alternates between two discrete levels of stiffness depending on the amplitude of the commanded force. The algorithm features a switching force-tracking control based on the forward-propagating Riccati equation. A disturbance-rejection component increases the device’s transparency in zero assistance mode. The force control was used to implement an assistive strategy that aims to correct the asymmetric gait typical of stroke survivors. Assistive joint torques synchronize with the user’s gait by means of an adaptive frequency oscillator, which extracts the continuous phase and frequency of the patient’s gait using data from both the paretic and the healthy sides. The control was tested with healthy subjects wearing the exoskeleton while subject to a simulated knee flexion impairment. The control proved effective in restoring spatial and temporal knee flexion symmetry to levels comparable to unobstructed gait.

(2021) Discrete Windowed-Energy Variable Structure Passivity Signature Control for Physical Human-(Tele)Robot Interaction, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2021.3064204.
Keywords: Algorithms; Biomechanics; Delay; Delays; Force; Haptics; Iterative methods; Passivity; physical human-robot interaction; Power system stability; Rehabilitation; Robots; Stability analysis; Telerobotics; Telesurgery; Variable structure control; variable structure passivity control.
Abstract: In this letter, we propose a novel adaptive iterative stabilization method for physical human-(tele)robot interaction, named Discrete Windowed-Energy Variable Structure Passivity Signature Control (DWE-VSPSC). The proposed stabilizer is capable of adaptively translating the knowledge domain regarding the capacity of the user’s biomechanics in absorbing physical interaction energy to reduce the transparency distortion (induced for stabilization) while enhancing system performance through a flexible design of stabilizer. The contributions of this letter are: (1) the design of a novel discrete adaptive stabilizer with proof of stability; allowing for digital implementation of the intelligent algorithm; (2) introducing the concept of “windowed energy stabilization” for the proposed variable structure passivity-signature controller in order to allow for tuning the energy behavior of the interconnected system while making a balance between conservatism and agility of the system; (3) relaxing any assumption on the passivity behavior of the environment while being able to handle stochastic variable network delays without any restriction on the rate of change of delay or the delay. The mathematical design of the stabilizer is provided alongside the stability proof. Also, the performance of the system is evaluated using a systematically-designed grid simulation study for a large range of delay and environmental impedances ranging from passive to non-passive behaviors. A direct application of the outcome is for telerobotic rehabilitation and telerobotic surgery.

(2021) Physical Human-Robot Collaboration: Robotic Systems, Learning Methods, Collaborative Strategies, Sensors, and Actuators, in IEEE T. Cybern., DOI: 10.1109/TCYB.2019.2947532.
Keywords: Actuators; Algorithms; Collaboration; collaborative robots; Control algorithms; human–robot collaboration (HRC); human–robot interaction; Machine learning; physical HRC (pHRC); Robot kinematics; Robot sensing systems; robotic systems; Robotics; Robots; Sensors; Task analysis.
Abstract: This article presents a state-of-the-art survey on the robotic systems, sensors, actuators, and collaborative strategies for physical human-robot collaboration (pHRC). This article starts with an overview of some robotic systems with cutting-edge technologies (sensors and actuators) suitable for pHRC operations and the intelligent assist devices employed in pHRC. Sensors being among the essential components to establish communication between a human and a robotic system are surveyed. The sensor supplies the signal needed to drive the robotic actuators. The survey reveals that the design of new generation collaborative robots and other intelligent robotic systems has paved the way for sophisticated learning techniques and control algorithms to be deployed in pHRC. Furthermore, it revealed the relevant components needed to be considered for effective pHRC to be accomplished. Finally, a discussion of the major advances is made, some research directions, and future challenges are presented.

(2021) Development of a Planar Haptic Robot With Minimized Impedance, in IEEE. Trans. Biomed. Eng., DOI: 10.1109/TBME.2020.3038896.
Keywords: Robots; End effectors; Haptic interfaces; Impedance; Actuators; Robot kinematics; Motor drives; User workspace; haptic robot; neurological disorder; upper limb; impedance.
Abstract: Several studies have reported that stroke survivors displayed improved voluntary planar movements when forces supporting the upper limb increased, and when impeding forces decreased. Earlier haptic devices interacting with the human upper limb were potentially impacted by undesired residual friction force and device inertia. To explore natural, undisturbed voluntary motor control in stroke survivors, we describe the development of a Decoupled-Operational space Robot for wide Impedance Switching (DORIS) with minimized mechanical impedances. This design is based on a novel decoupling mechanism separating the end effector from a manipulator. While the user manipulates the end effector freely inside the workspace of the decoupling mechanism, to which a manipulator of the robot is attached, the robot detects such change in position using a lightweight linkage system. The manipulator of the robot then follows such movements of the end effector swiftly. Consequently, the user can explore the extended workspace, which can be as large as the manipulator’s workspace. Since the end effector is mechanically decoupled from the manipulators and actuators, the user can remain unaffected by the mechanical impedances of the manipulator. Mechanical impedances perceived by the user and bandwidth of the control system were estimated. The developed robot was capable of detecting larger maximum acceleration and larger jerk of the reaching movement in chronic stroke survivors with hemiparesis. We propose that this device can be utilized for evaluating voluntary motor control of the upper limb while minimizing the impact of robot inertia and friction forces on limb behavior.

(2021) Split-Crank Functional Electrical Stimulation Cycling: An Adapting Admitting Rehabilitation Robot, in IEEE Trans. Control Syst. Technol., DOI: 10.1109/TCST.2020.3032474.
Keywords: Adaptation; Admittance; Asymmetry; Control stability; Cycles; Disorders; Electrical impedance; functional electrical stimulation (FES); Lyapunov; Lyapunov methods; Movement disorders; Muscles; Neuromuscular stimulation; nonlinear control; Nonlinear control systems; Passenger safety; Pedals; Rehabilitation; Rehabilitation robotics; Rehabilitation robots; Stability analysis; Stimulation; Systems stability; Tracking errors.
Abstract: Motorized functional electrical stimulation (FES) cycling is a promising rehabilitation strategy for individuals with movement disorders, particularly when the pedals of the FES cycle are decoupled to measure and address asymmetries. In this article, a rehabilitation robot, i.e., a split-crank FES cycle, is developed which utilizes a combined admittance-cadence controller to address rider asymmetries through adaptation, ensure rider safety, and electrically stimulate the rider’s leg muscles to pedal the cycle at the desired cadence. The theoretical development of the controllers is based on a combined Lyapunov-passivity switched systems stability analysis. Experiments were conducted on one able-bodied participant and three participants with various movement disorders, resulting in an average admittance tracking error of −0.13 ± 1.77 RPM with adaptation and −0.03 ± 4.05 RPM without adaptation. The split-crank FES cycle successfully admits to the rider, preserves rider safety, and offers a promising robotic rehabilitation strategy for individuals affected by movement disorders.

(2021) HapFIC: An Adaptive Force/Position Controller for Safe Environment Interaction in Articulated Systems, in IEEE Trans. Neural Syst. Rehabil. Eng., DOI: 10.1109/TNSRE.2021.3098062.
Keywords: Impedance; Haptic interfaces; Robots; Force; Task analysis; Robot sensing systems; Fractals; Haptics; force/position control; human–robot interaction.
Abstract: Haptic interaction is essential for the dynamic dexterity of animals, which seamlessly switch from an impedance to an admittance behaviour using the force feedback from their proprioception. However, this ability is extremely challenging to reproduce in robots, especially when dealing with complex interaction dynamics, distributed contacts, and contact switching. Current model-based controllers require accurate interaction modelling to account for contacts and stabilise the interaction. In this manuscript, we propose an adaptive force/position controller that exploits the fractal impedance controller’s passivity and non-linearity to execute a finite search algorithm using the force feedback signal from the sensor at the end-effector. The method is computationally inexpensive, opening the possibility to deal with distributed contacts in the future. We evaluated the architecture in physics simulation and showed that the controller can robustly control the interaction with objects of different dynamics without violating the maximum allowable target forces or causing numerical instability even for very rigid objects. The proposed controller can also autonomously deal with contact switching and may find application in multiple fields such as legged locomotion, rehabilitation and assistive robotics.

(2021) Variable Damping Control for pHRI: Considering Stability, Agility, and Human Effort in Controlling Human Interactive Robots, in IEEE T. Hum.-Mach. Syst., DOI: 10.1109/THMS.2021.3090064.
Keywords: Damping; Human-robot interaction; Impedance; Extremities; Stability; Tuning; Agility; human effort; performance; physical human robot interaction; stability; variable impedance control.
Abstract: This article presents a multi-degree-of-freedom variable damping controller to manage the trade-off between stability and agility and to reduce user effort in physical human-robot interaction. The controller accounts for the human body’s inherent impedance properties and applies a range of robotic damping from negative (energy injection) to positive (energy dissipation) values based on the user’s intent of motion. To evaluate the effectiveness of the proposed controller in balancing the trade-off between stability/agility and reducing user effort, two studies are performed on both the human upper-extremity and lower-extremity to represent both industrial and rehabilitation applications of the proposed controller. These studies required subjects to perform a series of multidimensional target reaching tasks while the human user interacted with either the end-effector of a robotic arm for the upper-extremity study or a wearable ankle robot for the lower-extremity study. Stability, agility, and user effort are quantified by a variety of performance metrics. Stability is quantified by both overshoot and stabilization time. Mean and maximum speed are used to quantify agility. To quantify the user effort, both overall and maximum muscle activation, and mean and maximum root-mean-squared interaction force are calculated. The results of both the upper- and lower-extremity studies demonstrate that the controller is able to reduce user effort while increasing agility at a negligible cost to stability.

(2021) A Framework of Hybrid Force/Motion Skills Learning for Robots, in IEEE Trans. Cogn. Dev. Syst., DOI: 10.1109/TCDS.2020.2968056.
Keywords: Computer programming; Contact force; Design factors; Dynamic movement primitives (DMPs); Dynamics; Force; force observer; Force sensors; generalization; Human factors; Human motion; hybrid force/motion control; Motion control; Muscles; Observers; Robot arms; Robot dynamics; Robot learning; Robot sensing systems; Robotics; Robots; skill transfer; Skills; Stiffness; Task analysis.
Abstract: Human factors and human-centered design philosophy are highly desired in today’s robotics applications such as human–robot interaction (HRI). Several studies showed that endowing robots of human-like interaction skills can not only make them more likeable but also improve their performance. In particular, skill transfer by imitation learning can increase the usability and acceptability of robots by users without computer programming skills. In fact, besides positional information, muscle stiffness of the human arm and contact force with the environment also play important roles in understanding and generating human-like manipulation behaviors for robots, e.g., in physical HRI and teleoperation. To this end, we present a novel robot learning framework based on dynamic movement primitives (DMPs), taking into consideration both the positional and contact force profiles for human–robot skills transferring. Distinguished from the conventional method involving only the motion information, the proposed framework combines two sets of DMPs, which are built to model the motion trajectory and the force variation of the robot manipulator, respectively. Thus, a hybrid force/motion control approach is taken to ensure the accurate tracking and reproduction of the desired positional and force motor skills. Meanwhile, in order to simplify the control system, a momentum-based force observer is applied to estimate the contact force instead of employing force sensors. To deploy the learned motion–force robot manipulation skills to a broader variety of tasks, the generalization of these DMP models in actual situations is also considered. Comparative experiments have been conducted using a Baxter robot to verify the effectiveness of the proposed learning framework on real-world scenarios like cleaning a table.

(2021) Plug-and-Train Robot (PLUTO) for Hand Rehabilitation: Design and Preliminary Evaluation, in IEEE Access, DOI: 10.1109/ACCESS.2021.3115580.
Keywords: Actuators; Caregivers; Computer & video games; Degrees of freedom; Ergonomics; Feedback; Forearm; Hand (anatomy); Hand rehabilitation; Pluto; Questionnaires; Rehabilitation; rehabilitation robot; Rehabilitation robotics; Robots; Thumb; Training; Usability; User experience; Wrist.
Abstract: Hand neurorehabilitation involves training movements at the forearm, wrist, fingers, and thumb joints. Assisted training of all these joints requires either one complex multiple degree-of-freedom (DOF) robot or a set of simple robots with one or two DOF. Neither of these is economic or clinically viable. This paper addresses this problem with the PLUg and train rObot (PLUTO)- a single DOF robot that can train multiple joints one at a time. PLUTO has a single actuator with a set of passive attachments/mechanisms that can be easily attached/detached to train for wrist flexion-extension, wrist ulnar-radial deviation, forearm pronation-supination, and gross hand opening-closing. The robot can provide training in active and assisted regimes. PLUTO is linked to performance adaptive computer games to provide feedback to the patients and motivate them during training. As the first step towards clinical validation, the device usability was evaluated by 45 potential stakeholders/end-users of the device, including 15 patients, 15 caregivers, and 15 clinicians with standardized questionnaires: System Usability Scale (SUS) and User Experience Questionnaire (UEQ). Patients and caregivers were administered the questionnaire after a two-session training with the robot. Clinicians, on the other hand, had a single session demo, after which feedback was obtained. The total SUS score obtained from patients, clinicians, and caregivers was 73.3 ± 14.6 (n = 45), indicating good usability. The UEQ score was rated positively in all subscales by both patients and clinicians, indicating that the features of PLUTO match their expectations. The positive response from the preliminary testing and the feedback from the stakeholders indicate that with additional passive mechanisms, assessment features, and optimized ergonomics, PLUTO will be a versatile, affordable, and useful system for hand rehabilitation.;Hand neurorehabilitation involves training movements at the forearm, wrist, fingers, and thumb joints. Assisted training of all these joints requires either one complex multiple degree-of-freedom (DOF) robot or a set of simple robots with one or two DOF. Neither of these is economic or clinically viable. This paper addresses this problem with the PLU g and train r O bot (PLUTO)- a single DOF robot that can train multiple joints one at a time. PLUTO has a single actuator with a set of passive attachments/mechanisms that can be easily attached/detached to train for wrist flexion-extension, wrist ulnar-radial deviation, forearm pronation-supination, and gross hand opening-closing. The robot can provide training in active and assisted regimes. PLUTO is linked to performance adaptive computer games to provide feedback to the patients and motivate them during training. As the first step towards clinical validation, the device usability was evaluated by 45 potential stakeholders/end-users of the device, including 15 patients, 15 caregivers, and 15 clinicians with standardized questionnaires: System Usability Scale (SUS) and User Experience Questionnaire (UEQ). Patients and caregivers were administered the questionnaire after a two-session training with the robot. Clinicians, on the other hand, had a single session demo, after which feedback was obtained. The total SUS score obtained from patients, clinicians, and caregivers was 73.3 ± 14.6 (n = 45), indicating good usability. The UEQ score was rated positively in all subscales by both patients and clinicians, indicating that the features of PLUTO match their expectations. The positive response from the preliminary testing and the feedback from the stakeholders indicate that with additional passive mechanisms, assessment features, and optimized ergonomics, PLUTO will be a versatile, affordable, and useful system for hand rehabilitation.

(2021) Human-Robot Cooperation Control Based on Trajectory Deformation Algorithm for a Lower Limb Rehabilitation Robot, in IEEE/ASME transactions on mechatronics, DOI: 10.1109/TMECH.2021.3053562.
Keywords: Algorithms; Compliance; Cooperation; Deformation; Electrical impedance; Human motion; Human-robot interaction; Human–robot cooperation control; Impedance; lower limb rehabilitation robot; Medical robotics; movement smoothness; Patient rehabilitation; Proportional derivative; Rehabilitation; Rehabilitation robotics; Rehabilitation robots; robot compliance; Robot control; Robots; Smoothness; Training data; Trajectory; Trajectory control; trajectory deformation algorithm (TDA); Trajectory planning.
Abstract: Although many studies have certified the advantages of human–robot cooperation control with admittance model (AM), robot compliance, and movement smoothness need to be further improved. In this article, a trajectory deformation algorithm (TDA) is developed as a high-level trajectory planner, which can plan subject’s desired trajectory based on interaction force during physical human–robot interaction (pHRI). A low-level proportional-derivative (PD) position controller is selected to ensure the lower limb rehabilitation robot can track the desired trajectory. Then, the validity of TDA is verified through simulation and experiment studies. The energy per unit distance (EPUD) and dimensionless squared jerk (DSJ) are chosen as indicators of robot compliance and movement smoothness, respectively. The experimental results demonstrated that both the EPUD and the DSJ values using TDA are smaller than that using the AM, indicating the TDA can improve robot compliance and movement smoothness. Therefore, it may have great potential in fields involving pHRI, such as robot-aided rehabilitation.

(2021) Development of an Intention-Based Adaptive Neural Cooperative Control Strategy for Upper-Limb Robotic Rehabilitation, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2020.3043197.
Keywords: Adaptive control; Adaptive filters; Biology; Control stability; Cooperative control; Force; Human engineering; Human motion; intention recognition; Kalman filter; Kalman filters; Muscles; Radial basis function; Rehabilitation; Rehabilitation robotics; Rehabilitation robots; Robot sensing systems; Robots; Robust control; Sliding mode control; Smoothness; Strategy; Training; Trajectory analysis.
Abstract: Robotic rehabilitation therapy has become an important technology to recover the motor ability of disabled individuals. Clinical studies indicate that involving the active intention of patient into rehabilitation training contributes to promoting the performance of therapies. An adaptive neural cooperative control strategy is developed in this letter to realize intention-based human-cooperative rehabilitation training. The human motion intention is estimated by fusing the human-robot interaction forces and the muscular forces into a Gaussian radial basis function network. A biological force estimation method is proposed to obtain the muscular forces of biceps and triceps based on surface electromyography signals and Kalman filter. A robust adaptive sliding mode controller is integrated into the cooperative control scheme to ensure the accuracy and stability of inner position control loop with uncertainties. The minimum jerk cost principle is used to improve the smoothness and continuity of trajectory. To evaluate the effectiveness of the proposed control scheme, further experimental investigations are conducted on a planar upper-limb rehabilitation robot with ten volunteers. The results indicate that the proposed control strategy has significant potential to modulate the interaction compliance and cooperation process during training.

(2021) Performance-Based Hybrid Control of a Cable-Driven Upper-Limb Rehabilitation Robot, in IEEE. Trans. Biomed. Eng., DOI: 10.1109/TBME.2020.3027823.
Keywords: Active damping; assist-as-needed control; cable-driven robot; Controllers; Damping; Disturbance; Force; Fuzzy logic; Hybrid control; Immune system; Motor task performance; Rehabilitation; Rehabilitation robotics; Rehabilitation robots; Robots; Stiffness coefficients; Stroke; Task analysis; Tracking errors; Training; upper-limb rehabilitation.
Abstract: Patients after stroke may have different rehabilitation needs due to various levels of disability. To satisfy such needs, a performance-based hybrid control is proposed for a cable-driven upper-limb rehabilitation robot (CDULRR). The controller includes three working modes, i.e., resistance mode, assistance mode and restriction mode, which are switched by the tracking error since it is a common index to represent motor performance. In resistance mode, the proper damping force would be provided for subjects, which is in the opposite direction to the actual velocity. In assistance mode, a method of adjusting stiffness coefficient by fuzzy logic is adopted to provide suitable assistance to help subjects. In restriction mode, the damping force is applied again to limit the movement and ensure the safety. To verify the effectiveness of the controller, the task-oriented experiments with different disturbance were conducted by ten healthy subjects. The experiments results demonstrated that the controller can adjust working modes by the subjects’ motor performance. It was found that, as the increasing disturbance led to a decrease in the motor performance, the robot provided more assistance in the trainings. Adaptive adjustment of damping force and stiffness coefficient allowed the controller to induce more active effort.

(2021) A Hybrid Arm-Hand Rehabilitation Robot With EMG-Based Admittance Controller, in IEEE Trans. Biomed. Circuits Syst., DOI: 10.1109/TBCAS.2021.3130090.
Keywords: Active control; Admittance control; Arm; cable-driven parallel mechanism; Controllers; Electrical impedance; Electromyography; EMG; Exoskeleton; Exoskeleton Device; Exoskeletons; Grasping (robotics); Hand; Hand (anatomy); hand exoskeleton; Humans; hybrid structure; Hybrid structures; reach-and-grasp; Rehabilitation; Rehabilitation robot; Rehabilitation robotics; Rehabilitation robots; Robotics; Stroke; Stroke Rehabilitation - methods; Three dimensional motion; Training.
Abstract: Reach-and-grasp is one of the most fundamental activities in daily life, while few rehabilitation robots provide integrated and active training of the arm and hand for patients after stroke to improve their mobility. In this study, a novel hybrid arm-hand rehabilitation robot (HAHRR) was built for the reach-and-grasp task. This hybrid structure consisted of a cable-driven module for three-dimensional arm motion and an exoskeleton for hand motion, which enabled assistance of the arm and hand simultaneously. To implement active compliance control, an EMG-based admittance controller was applied to the HAHRR. Experimental results showed that the HAHRR with the EMG-based admittance controller could not only assist the subject in fulfilling the reach-and-grasp task, but also generate smoother trajectories compared with the force-sensing-based admittance controller. These findings also suggested that the proposed approach might be applicable to post-stroke arm-hand rehabilitation training.

(2021) Voluntary Control of an Ankle Joint Exoskeleton by Able-Bodied Individuals and Stroke Survivors Using EMG-Based Admittance Control Scheme, in IEEE. Trans. Biomed. Eng., DOI: 10.1109/TBME.2020.3012296.
Keywords: Electromyography; Admittance; Exoskeletons; Torque; Stability analysis; Robots; Muscles; Admittance control; Human-robot cooperation control; Robot-assisted rehabilitation; Stroke; Surface electromyography (sEMG).
Abstract: Control schemes based on electromyography (EMG) have demonstrated their superiority in human-robot cooperation due to the fact that motion intention can be well estimated by EMG signals. However, there are several limitations due to the noisy nature of EMG signals and the inaccuracy of EMG-force/torque estimation, which might deteriorate the stability of human-robot cooperation movement. To improve the movement stability, an EMG-based admittance control scheme (EACS) was proposed, comprised of an EMG-driven musculoskeletal model (EDMM), an admittance filter and an inner position controller. To investigate the performance of EACS, a series of sinusoidal tracking tasks were conducted with 12 healthy participants and 4 stroke survivors in an ankle exoskeleton in comparison with the EMG-based open-loop control scheme (EOCS). The experimental results indicated that both EACS and EOCS could improve stroke survivors’ ankle range of motion (ROM). The experimental results of both healthy participants and stroke survivors showed that the assistance torque, tracking error and jerk values of EACS were lower than those of EOCS. The interaction torque of EACS decreased towards the increasing assistance ratio while that of EOCS increased. Moreover, the EMG levels of tibialis anterior (TA) decreased towards the increasing assistance ratio but were higher than those of EOCS. EACS was effective in improving movements stability, and had the potential to be applied in robot-assisted rehabilitation training to address the foot-drop problem.

(2021) Active Disturbance Rejection Control for a Fluid-Driven Hand Rehabilitation Device, in IEEE/ASME transactions on mechatronics, DOI: 10.1109/TMECH.2020.3006364.
Keywords: Active control; Active disturbance rejection control; Actuators; Algorithms; Control methods; Elastomers; Exoskeletons; extended state observer; extension motion; Fluid pressure; hand rehabilitation; IEEE transactions; Mechatronics; parameter selection; Parameters; Performance evaluation; Proportional integral; Rehabilitation; Rejection; State observers; steady-state performance; Training; transient.
Abstract: This article proposes a new hand rehabilitation device to help therapists repetitively perform hand group stretching training for poststroke patients. The actuator of the proposed device is made of elastomeric materials, driven by fluid pressure, and put onto the surface of a glove. Compared to the traditional rigid exoskeletons for hand rehabilitation, the proposed device is soft, lightweight, and low cost. Because the exact and explicit model is hard to be built for the proposed device and the unconscious tremors occurred during hand rehabilitation, the active disturbance rejection control (ADRC) algorithm, based on the extended state observer (ESO), is adopted to achieve the control purpose. This article analyzes the transient and steady-state performances of the controller based on some reasonable assumptions, and shows its better disturbance rejection ability compared to the widely used proportional-integral-differential control method. In addition, the ADRC algorithm with the reduced-order extended state observer (RESO) has been studied as well. The parameters of the RESO can be obtained by the parameters of the previously determined ESO instead of the trial and error approach, which can ensure the control performance of the ADRC with the RESO does not degrade. Finally, clinical tests have been conducted to verify the functional correctness of the proposed device, and these theoretical results have also been verified by experiments and comparisons.

(2021) Autonomy in Physical Human-Robot Interaction: A Brief Survey, in IEEE Robot. Autom. Lett., DOI: 10.1109/LRA.2021.3100603.
Keywords: Autonomy; Collaboration; Computer Science; Human engineering; human-centered robotics; human-robot collaboration; Human-robot interaction; Mobile robots; Physical human-robot interaction; Physical workload; Robot control; Robot kinematics; Robot sensing systems; Robotics; Robots; Task analysis.
Abstract: Sharing the control of a robotic system with an autonomous controller allows a human to reduce his/her cognitive and physical workload during the execution of a task. In recent years, the development of inference and learning techniques has widened the spectrum of applications of shared control (SC) approaches, leading to robotic systems that are capable of seamless adaptation of their autonomy level. In this perspective, shared autonomy (SA) can be defined as the design paradigm that enables this adapting behavior of the robotic system. This letter collects the latest results achieved by the research community in the field of SC and SA with special emphasis on physical human-robot interaction (pHRI). Architectures and methods developed for SC and SA are discussed throughout the letter, highlighting the key aspects of each methodology. A discussion about open issues concludes this letter.;Sharing the control of a robotic system with an autonomous controller allows a human to reduce his/her cognitive and physical workload during the execution of a task. In recent years, the development of inference and learning techniques has widened the spectrum of applications of shared control (SC) approaches, leading to robotic systems that are capable of seamless adaptation of their autonomy level. In this perspective, shared autonomy (SA) can be defined as the design paradigm that enables this adapting behavior of the robotic system. This letter collects the latest results achieved by the research community in the field of SC and SA with special emphasis on physical human-robot interaction (pHRI). Architectures and methods developed for SC and SA are discussed throughout the paper, highlighting the key aspects of each methodology. A discussion about open issues concludes this letter.

(2021) Satisfying Task Completion and Assist-as-Needed Performance in Robotic Exoskeletons, in IEEE transactions on medical robotics and bionics, DOI: 10.1109/TMRB.2021.3097132.
Keywords: Assist-as-needed control; Control methods; Control systems design; Controllers; disturbance observer; Disturbance observers; Dynamic models; Exoskeletons; Impedance; impedance control; Rehabilitation robotics; Rehabilitation robots; Robot control; Robot sensing systems; Robotics; Robots; Task analysis; Tracking control; upper-limb exoskeleton; Velocity; Velocity measurement.
Abstract: Assist-as-needed (AAN) control approaches aim to minimize the intervention of robots in robot-aided rehabilitation therapy when patients are able to perform the task. The existing approaches generally contain a forgiving controller which fails to satisfactorily complete the task when patients stay passive or apply resistive/opposing forces. Task completion in such situations is important since it efficiently trains and guides users on how to perform the exercise during active participation. In this study, we have developed an AAN control scheme that provides significant freedom to users when they can perform the exercise while it provides sufficient support for the precise completion of the task when users are not able to do so. These features have been achieved by modulating the impedance of the robot and designing the controller in the velocity domain. A new velocity tracking controller is proposed which does not utilize velocity measurements, and two methods are considered to measure human inputs: force sensor-based method and disturbance observer-based method. In the first approach, the controller does not require the dynamics of the human-robot system, and in the second approach, partial information of the dynamics is required. The performance of the controller is evaluated on an upper-limb robotic exoskeleton. Note to Practitioners -The classic control methods utilized in rehabilitation robots for the active participation of patients are usually low-gain controllers that allow patients to deviate from the desired motion path. However, this approach results in poor task completion when users are not able to do the task. This paper proposes a control scheme for robotic exoskeletons that simultaneously satisfies the two objectives, namely, the low intervention of the robot when users can perform the task, and precise task completion when users cannot handle it. The proposed controller does not require velocity measurements and the precise dynamic model of the system. The approach needs human intention detection, which can be achieved by using force sensors or a disturbance observer. While the second method reduces the complexity and cost of the mechanical parts, it requires the dynamic model of the robot. The tests conducted on an upper-limb robotic exoskeleton show the satisfactory performance of both methods.