Robust Self-Organizing Neural-Fuzzy Control With Uncertainty Observer for MIMO Nonlinear Systems

Abstract

This paper proposes a robust self-organizing neural-fuzzy-control (RSONFC) scheme for a class of uncertain nonlinear multiple-input-multiple-output (MIMO) systems. We first develop a self-organizing neural-fuzzy network (SONFN) with concurrent structure and parameter learning. The fuzzy rules of SONFN are generated or pruned systematically. The proposed RSONFC scheme comprises an SONFN identifier, an uncertainty observer, and a supervisory controller. The SONFN identifier functions as the principal controller, and the uncertainty observer is designed to oversee uncertainties within the compound system. The supervisory controller combines sliding-mode control (SMC) and an adaptive bound-estimation scheme with various weights to achieve H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> tracking performance with a desired level of attenuation. Projection-type adaptation laws of network parameters developed using the Lyapunov's synthesis approach guarantee the stability of the overall control system. Simulation studies on a single-link flexible-joint manipulator and a two-link robot demonstrate the effectiveness of the proposed control scheme.