Robust PI Control for Lower Limb Exoskeleton Robot Based on the Moth Flame Optimization Algorithm

Abstract

The purpose of this paper is to apply an intelligent algorithm to conduct the torque tracking control for lower limb exoskeleton robot driven by an electro-hydraulic servo system (EHSS). The dynamics of EHSS actuating the robot are mathematically modeled with two degrees of freedom joints, and the torque control strategy is made to realize the minimization of human-machine forces. The PI controller is selected to implement this strategy, and the selection of PI gains is important for system control. Therefore, the ameliorative moth flame optimization (AMFO) algorithm is chosen to optimize the PI gains. The main idea of moth flame optimization is to mimic the evolution of a moth’s lateral positioning mechanism over time. The AMFO algorithm is capable of achieving enhanced better global and local search capabilities by adding the inertia weights to the position update formula. Moreover, the particle swarm optimization (PSO) and whale optimization algorithm (WOA) are selected to obtain comparative results. The results indicate that, when compared to the WOA-PI and PSO-PI controllers, the AMFO-PI controller gains the least amount in terms of rise time, overshoot, steady error, mean absolute error, and root mean square error.