Intelligent Controllers which can Adapt to Changing Environments: Anticipatory Neuro-Fuzzy Control
Claire Lynne McCullough · 1993
An anticipatory neuro-fuzzy controller has been developed which combines traditional fuzzy logic with feedback concepts and a neural network to predict system behavior. Fuzzy rules in the system are 'nested' so that it is not necessary to evaluate all rules at each time step if some smaller subset of rules will produce acceptable system behavior. Anticipatory neuro-fuzzy control does not require that any mathematical model exist for the system to be controlled, and can therefore be applied to many real world systems for which other, more traditional control methods cannot be used.