Robust Adaptive Tracking Control of Uncertain Rehabilitation Exoskeleton Robot

Rehabilitation robots have become an influential tool in physiotherapy treatment because they are able to provide intensive rehabilitation treatment over a long period of time. However, this technology still suffers from various problems such as dynamic uncertainties, external disturbances, and human–robot interaction. In this paper, we propose a robust adaptive control approach of an exoskeleton robot with an unknown dynamic model and external disturbances. First, the dynamics of the exoskeleton's arm is presented. Then, we design a robust adaptive sliding mode control in which the parameter uncertainties and the disturbances are estimated by the adaptive update methods. The proposed control ensures a good tracking of the system with a finite time convergence of sliding surface to zero. Throughout this paper, the designed control law and the global stability analysis are formulated and demonstrated based on the appropriate choice of the candidate Lyapunov function. The experimental and comparative results, performed for seven degrees-of-freedom (DOFs) exoskeleton arm with three healthy subjects to track a passive rehabilitation motion, confirm the effectiveness and robustness of the proposed control law compared with conventional adaptive approach.