Robust controller design for lower limbs rehabilitative training robot

Abstract

A lower limbs rehabilitative training robot was designed to improve walking ability of patients who suffer from impairment in locomotion after neurology injuries or aged persons who lack walking capability. The robot was fixed with four omni-directional wheels which can move within any radius in any direction. When the robot is running, some undetected uncertainty disturbances change its running trajectory, if the controller does not consider these factors. An L2 gain robust controller was designed to solve undetected uncertainty disturbance restraint problem. Backstepping scheme was employed to deduce robust control law. Simulation experiment results show that the controller is effective. ©2010 IEEE.