Representing and optimizing fuzzy-controllers by neural networks

A couple of different methods is known for combining fuzzy-controllers with neural networks. One of the reasons for these combinations is to work around the fuzzy-controllers´ disadvantage of not being adaptive. Therefore, it is helpful to represent a given fuzzy-controller by means of a neural network and to have the rules adapted by a special learning algorithm. Some of these methods are applied to the NEFCON-model or the model of Lin and Lee (1994). However, none of these methods is able to adapt all fuzzy-controller components. In this paper we suggest a new model, which gives the user the ability to represent a given fuzzy-controller by a neural network and adapt all of its components as desired.