Model based adaptive controller with grasshopper optimization algorithm for upper-limb rehabilitation robot

Abstract

Model based adaptive controllers (MBACs) are considered one of the most common adaptive controllers that are used with robotic systems due to their ensuring nonlinear robust scheme with global asymptotic stability for controlling nonlinear systems. However, this controller requires precise mathematical models of the controlled systems. In this paper, an optimal model-based adaptive controller (OMBAC) is suggested for controlling a two-link upper limb rehabilitation robot. This controller, in the presence of model uncertainties, can guarantee the robustness of the rehabilitation robot. Although the OMBAC is an adaptive and model-based controller, some of its parameters need to be determined precisely. In this paper, these parameters are determined by the grasshopper optimization algorithm (GOA). The Lyapunov method is used to analyze the stability assurance of controlled rehabilitation. The results of the simulation for two tested trajectories (linear and nonlinear trajectories) demonstrate the efficiency of the suggested OMBAC with fast settling time, minimum error steady state, and very small overshoot.