Fuzzy Logic Normal Forms for Control Law Representation

Abstract

The main paradigm of fuzzy control is that the control law is formulated as a knowledge based algorithm expressed in the language of logical formulas with vague predicates. It allows us to express qualitative characterization of variables using fuzzy predicates, and functional dependencies between variables using conditional sentences with fuzzy predicates (rules). Such an approach makes it possible to implement human reasoning in a form of a sequence of rules in the control algorithm. This fundamental idea introduced by L.A. Zadeh brought a new fruitful methodology to control theory. A lot of examples can be found where expert knowledge can be represented in the form of a rule based control algorithm. It has been demonstrated in practice that fuzzy control can be efficiently used to control complex systems both technical as well as those in which the human element plays a significant role. To clearly understand what the main advantages of the fuzzy approach to control are, let us compare it with the conventional control techniques.