Observer-based fuzzy adaptive control for a class of nonlinear systems: real-time implementation for a robot wrist

This paper examines the tracking control problem for a class of feedback linearizable nonlinear systems for which there is no available analytic model. Based on the ability of fuzzy systems to approximate any nonlinear mapping, the unknown nonlinear system is represented by a Takagi-Sugeno (TS) fuzzy system. First, we represent the nonlinear plant with an adaptive TS fuzzy system, where the parameters are adjusted via adaptive laws according to the Lyapunov and passivity theories. Then, a fuzzy adaptive feedback linearizing controller is designed under the constraint that only the output of the plant is available for measurement. According to this constraint, a plant state observer is introduced in the proposed control structure. Finally, the feasibility of the proposed methodology is experimentally validated with a real-time implementation.

I. Introduction

Today, advancing technology and competitive markets require industrial processes to operate in many situations where linear approximation of nonlinear plants is no longer satisfactory. Using this approach, the performance of the linear controller is guaranteed to be optimal only in the neighborhood of the linearization point.