Passive Hand Rehabilitation Training Through Robots: an Iterative Learning Control Approach

Abstract

Robots are widely used in the field of medical rehabilitation to assist patients to conduct rehabilitation training. Due to the repetition nature of the rehabilitation training, this paper proposes an iterative learning controller equipped with feedback mechanism for the passive hand rehabilitation training, which is based on a cable-driven hand exoskeleton robot. Thanks to the use of the information from previous iterations, the desired trajectory can be perfectly tracked over a finite duration. Moreover, the monotonic convergence of the proposed iterative learning controller can be achieved under a sufficient condition. In addition, our iterative learning controller is applied to the hand exoskeleton robot in both the simulation tests and the physical trajectory tracking experiments, which demonstrates its effectiveness and desired control performance.