An approach to the design of a fuzzy self-tuning PID controller

Abstract

This paper presents an approach to the design of a self-tuning PID controller using a fuzzy adaptive mechanism. The main idea is to tune the controller gains on-line by a fuzzy adaptive mechanism, i.e. the controller gains are subjected to a single parameter that results from a fuzzy inference mechanism. The controller is simulated with a 2-DOF robot and simulations show a considerable improvement in performance, in terms of setpoint responses, as compared to conventional well-tuned PID controllers.