Design of generalized Sugeno controllers by approximating hybrid fuzzy-PID controllers

Abstract

This paper proposes a simple technique to design a generalized Sugeno-type controller (GSC). A hybrid fuzzy-PID (HFPID) controller is approximated using recursive least squares by a Sugeno-type controller. The method used to generate the approximating data is shown to affect the approximation. A comparison of the performance between a Sugeno-type controller and a hybrid fuzzy-PID controller when they are both applied to a two-degree-of-freedom robot manipulator arm is shown. Furthermore, the number of terms needed to produce accurate performance of the generalized Sugeno-type controller is also discussed.