powered by:
MagicWare, s.r.o.

Composite Adaptive Fuzzy Control

Authors:Bellomo Domenico, Politecnico di Bari, Italy
Naso David, Politecnico di Bari, Italy
Turchiano Biagio, Politecnico di Bari, Italy
Babuska Robert, Delft University of Technology, Netherlands
Topic:3.2 Cognition and Control ( AI, Fuzzy, Neuro, Evolut.Comp.)
Session:Adaptive Neuro-fuzzy Control
Keywords: Adaptive control, fuzzy systems, model reference control, feedback linearization,Lyapunov stability, parameter estimation

Abstract

Adaptive fuzzy control has been a topic of active research over the last decade.However, most efforts have been directed toward one goal: achieving asymptoticstability and tracking. Little attention has been paid to the accuracy of theidentified fuzzy models and to their transparency and interpretability whereas theseshould be the key aspects motivating the use of fuzzy models in adaptive control.The main contribution of this paper is to present an adaptive fuzzy controller withcomposite adaptive laws based on both tracking and prediction error. Compared toother adaptive fuzzy controllers, the proposed controller achieves smootherparameter adaptation, better accuracy and improved performance. It overcomes some ofthe drawbacks of similar schemes described in the literature on adaptive fuzzycontrol. The limitations of the proposed approach are also discussed.