Composite Adaptive Fuzzy Control

Adaptive fuzzy control has been a topic of active research over the last decade. However, most efforts have been directed toward one goal: achieving asymptotic stability and tracking. Little attention has been paid to the accuracy of the identified fuzzy models and to their transparency and interpretability whereas these should 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 with composite adaptive laws based on both tracking and prediction error. Compared to other adaptive fuzzy controllers, the proposed controller achieves smoother parameter adaptation, better accuracy and improved performance. It overcomes some of the drawbacks of similar schemes described in the literature on adaptive fuzzy control. The limitations of the proposed approach are also discussed. Copyright c © 2005 IFAC