Optimal Tuning of an Active Disturbance Rejection Controller Using a Particle Swarm Optimization Algorithm

Abstract

This work deals with the optimal tuning of an Active Disturbance Rejection Controller (ADRC), which is composed of a Luenberger and a Disturbance Observers. The ADRC is applied to the position control of a servo system composed of a DC motor and its associated electronics. The goal of this controller is to reject the disturbances affecting the servo system and to impose a desired closed-loop dynamics. The Luenberger Observer (LO) provides angular velocity estimates whereas the Disturbance observer generates an estimate of the disturbance. The tuning of the ADRC to obtain optimal performance is not an easy task since the theoretical results associated to this controller only focuses on the stability of the closed-loop system. Therefore, three Particle Swarm Optimization (PSO) algorithms are implemented to tune the gains of the ADRC. The restrictions imposed to the particles in the PSO are obtained from the stability analysis of the ADRC. In this way, the PSO produces particles associated to optimal gains that guarantees closed-loop stability and at the same time the minimization of the Fitness Function. Real-time experiments are shown to assess the performance of the ADRC under the PSO tuning.