Robust multi-controller design with adaptive fuzzy switching system to robot wrist

Abstract

This paper presents a multi-controller approach of control with symbolic fuzzy switching system and adaptive fuzzy gain to control our process in the different environment. First we present the model of the process (wrist of a Stäubli robot RX 90) and a local parametric models around equilibrium point. Then, we introduce the principles of multi-controller approach and fuzzy switching system. The originality of our approach lies in the use of an integrator in the process to avoid the use of an explicit equilibrium point in the control law and blend the control signals provided by several controllers, which are model reference output feedback controllers (RST controllers) and the switching system will use the information to blend some selected control signals. The weights of the local control law are determined from the quadratic criteria from fixed supervisory system using a fuzzy system. In order to show the efficiency of the proposed method, some 3D simulation results in CAO solid Works software are given.