Adaptive Tracking Control Based on GFHM for a Reconfigurable Lower Limb Exoskeleton

Abstract

Paraplegic patients can gain the ability of walking, sitting, standing, ascending stairs and slopes and vice versa by wearing lower limb exoskeletons. As a result of the direct contact between the human and robot, challenges such as unknown disturbances and dynamic uncertainties, which further emphasizes the performance of the control, arise. In this paper, an adaptive tracking control based on Generalized Fuzzy Hyperbolic Model (A-GFHM), a simple structured controller with low computational cost, is proposed for a lower limb exoskeleton. The proposed controller has proven to be highly capable of controlling MIMO non-linear systems with the challenges mentioned. Finally, the proposed real-time method is tested by experimental implementation on an exoskeleton robot manufactured at the Robotic Laboratory of the Ferdowsi University of Mashhad (FUM-Exoskeleton or FUME). By comparing the results with a PID controller, it becomes clear that the proposed method reduces the tracking error by 33.56% and 77.73%, in controlling the knee and hip joints, respectively.