Fuzzy-adaptive Impedance Control of Upper Limb Rehabilitation Robot Based on sEMG

Abstract

In most rehabilitation robot control strategies, voluntary residual motor function of the stroke patients’ affected limb have not been involved enough. In this paper, we propose a novel fuzzy-adaptive impedance control method based on sEMG. First, muscle activation is calculated online to estimate the voluntary residual motor function of the human upper limb. Then, an impedance controller and a PID force regulator are constructed to control position and interactive force between human upper limb and robot. Moreover, the robot's impedance control parameters are regulated in real time according to the muscle activation by using fuzzy inference system. At last, we apply this control strategy to an upper limb rehabilitation robot. The experiment results show robot can regulate the actual assistive force and impedance in real time based on voluntary participation intention of the human, which indicates it is effective for the proposed impedance control strategy to be an "assist-as-needed" strategy.