Neural learning impedance control of lower limb rehabilitation exoskeleton with flexible joints in the presence of input constraints

This article presents neural learning based adaptive impedance control for a lower limb rehabilitation exoskeleton with flexible joints (LLREFJ). First, the full model consisting of both the rigid link and the flexible joint is obtained for the LLREFJ. Second, neural networks are used to compensate for the system uncertainties and external disturbance and an adaptive impedance controller is proposed by establishing an impedance error. In order to improve the control performance and enhance the system robustness, periodic dynamics is considered according to the repetitive motion of the rehabilitation process and handled by a repetitive learning algorithm. Then, the stability of the full system is proved rigorously by Lyapunov methods. Finally, comparative simulation reveals that the designed adaptive neural learning controller has improved the control performance.