An Iterative Algorithm for Fuzzy Modeling and Controller Design

Abstract

This paper presents an iterative LMI approach to fuzzy modeling and controller design for nonlinear systems with unknown parameters. The main feature is to achieve iteratively and effectively the fuzzy modeling and controller design. Three LMI conditions to identify the parameters of Takagi-Sugeno fuzzy models are derived. The parameters are identified by solving a minimization problem utilizing the LMIs. More importantly, we design a robust fuzzy controller that compensates the identification error occurred in the fuzzy modeling part. A design example shows the utility of the iterative approach to the fuzzy modeling and controller design.