Service Robotics: Robot-Assisted Training for Stroke Rehabilitation

Abstract

Significant motor improvements in the poststroke subjects were obtained after the elbow training assisted by the cEMG-driven robot. Robotic parameters also could be used to quantitatively monitor the process of the motor recovery, which provided more information on the extent of the motor recovery than the clinical scores assessed before and after the training. In the future works, comparison should be carried out between the training effects by the cEMG-driven robot and those by the other rehabilitation robots with control strategies, such as, the EMG-triggered control 15, impedance control 11, etc, by large-scale randomized control trials. In the robot design aspect, the cEMG-driven interactive control algorithm can be applied to other joints, such as, the wrist, the knee, and the ankle. The related robot-assisted training effects on the different parts of the body also need to be evaluated in future works. It is possible that the training effects may be elevated by integrating more vigorous visual sensory feedbacks, such as virtual reality programs.