A new methodology of fuzzy constraint-based controller design via constraint-network processing

Abstract

A complete design framework for a constraint-based controller based on fuzzy-constraint processing and its semantics and relationship to logic is presented. In this paper, the concept of fuzzy in problem solving is introduced, and some basic definitions of fuzzy-constraint processing in a constraint network and its semantic modeling are addressed. Then a local propagation inference mechanism for reasoning about imprecise information applying the filter operation in a network of constraints is proposed. Moreover, we advance the concurrent fuzzy-logic controller (FLC) to a new type of controller, the constraint-based controller (FCC), using a more general predicate calculus and full first-order logic knowledge representation and making use of the idea of fuzzy-constraint processing to model practical dynamic control systems. Finally, simulation results show that a FCC achieves equivalent performance as PD type and PI type FLCs and it also demonstrates superior outcomes to a conventional PID controller in terms of rise time and peak-percent overshoot.