Abstract:People with movement disorders are plagued with debilitating conditions, which significantly degrade their quality of life. Traditional rehabilitation typically involves intensive interaction between patients and therapists. While effective, traditional rehabilitation cannot keep abreast of the increasing patient population primarily attributed to a higher surviving rate after diseases and/or injuries. Furthermore, patients living in the rural areas have fairly limited access to rehabilitation services. In the past two decades, tremendous efforts have been put into developing rehabilitation and assistive robots to facilitate the rehabilitation training while relieving the physical involvement of therapists and/or lowering the related cost. Most notably, the rehabilitation and assistive robots have been significantly advanced with developments in actuators, sensors, microprocessors, and mobile software platforms. However, unlike traditional robotics, the intimate interaction between robot and human in rehabilitation robots indicates that the success is also closely related to a thorough understanding of the human neuromuscular aspects and human-machine interaction. This special issue primarily aims to gather the latest achievements in rehabilitation robots, exoskeletons, and prostheses including the following topics: (a) development of rehabilitation robots, exoskeleton, and upper/lower limb prostheses driven by bionics; (b) functional evaluation of rehabilitation robots, exoskeleton, and upper/lower limb prostheses with an emphasis on human movement biomechanics; (c) musculoskeletal modeling and simulation of human movements while wearing exoskeleton or prostheses; (d) noninvasive human-machine interface based on electromyography and/or electroencephalogram; (e) sensors for monitoring kinematics/kinetics, as well as biological signals in real time; (f) innovative actuators and control algorithms applied to rehabilitation robots, exoskeletons, and prostheses. In this special issue, collective studies address the aforementioned key elements via both technical and biomechanical approaches. A reconfigurable robotic hand exoskeleton was proposed to meet the fast growing need in hand rehabilitation. A novel control algorithm integrating sliding model control with cerebellar model articulation controller neural network was implemented in lower limb exoskeleton to enhance the coordination between patient and exoskeleton. An upper limb exoskeleton was enhanced with integrated optical cameras to offer more accurate estimation of joint posture than traditional motion capture system. A hybrid upper limb rehabilitation system consisting of a shoulder-elbow-forearm exoskeleton and a robotic manipulator was validated and tested in the clinic. The characteristics of muscle-tendon stimulation such as perception threshold and vibration frequency significantly influenced the muscle forces as well as the reaction time. Patellar retention was found to be superior to patellar replacement in knee arthroplasty via a comprehensive computer simulation. These collective studies, as part of the latest representative work, offered some new insights into the development and implementation of rehabilitation and assistive robots. Fan Gao Guanglin Li Huapeng Wu Qining Wang Jie Liu Justin Keogh

Cooperative Markov Decision Process model for human–machine co-adaptation in robot-assisted rehabilitation

Human Machine Co-adaption Interface via Cooperation Markov Decision Process System

Human-Machine Co-Adaptation for Robot-Assisted Rehabilitation via Dual-Agent Multiple Model Reinforcement Learning (DAMMRL)

Model Predictive Control for Human-Centred Lower Limb Robotic Assistance

Adaptive Learning based Upper-Limb Rehabilitation Training System with Collaborative Robot

Integrating subject-specific workspace constraint and performance-based control strategy in robot-assisted rehabilitation

Temporal-Difference Learning to Assist Human Decision Making during the Control of an Artificial Limb

Employing Socially Interactive Agents for Robotic Neurorehabilitation Training

Performance-based iterative learning control for task-oriented rehabilitation: a pilot study in robot-assisted bilateral training

Multi-Sensor Fusion-Based Mirror Adaptive Assist-as-Needed Control Strategy of a Soft Exoskeleton for Upper Limb Rehabilitation

Multi-mode adaptive control strategy for a lower limb rehabilitation robot

Advances in Rehabilitation and Assistive Robots for Restoring Limb Function in Persons with Movement Disorders.

AI enhanced collaborative human-machine interactions for home-based telerehabilitation

Assist-As-Needed Control Strategy of Bilateral Upper Limb Rehabilitation Robot Based on GMM

A co-design approach for a rehabilitation robot coach for physical rehabilitation based on the error classification of motion errors

Game-Theoretic Modeling of Human Adaptation in Human-Robot Collaboration

Iterative Learning-Based Path Control for Robot-Assisted Upper-Limb Rehabilitation

Socially Interactive Agents for Robotic Neurorehabilitation Training: Conceptualization and Proof-of-concept Study

Joint Action is a Framework for Understanding Partnerships Between Humans and Upper Limb Prostheses

Adaptive control for shape memory alloy actuated systems with applications to human–robot interaction

EEG-Based Analysis of Brain Responses in Multi-Modal Human-Robot Interaction: Modulating Engagement