Optimized Control Design of LQR for Flexible Joint Manipulator

Abstract

In this work, considering some important indices such as speed of the system response, control effort with minimum error and vibrations of the flexible joint manipulator system combining them into an objective function, an optimization problem is defined in order to find the optimal weighting matrices Q and R in LQR controller. To solve this optimization problem, intelligent optimization technique such as Particle Swarm Optimization (PSO) is used. PSO is applied to flexible joint manipulator to enhance the performance of the controller and optimally adjusting the weighting matrices Q and R matrix in LQR controller. The response obtained for manually tuned by hit and trial and LQR tuned by PSO are compared in terms of time domain characteristics. Effectiveness and optimal response of the proposed controller are achieved on Matlab/simulink.