Fuzzy and neural networks controller
Miguel Strefezza, Y. Dote · 2002
The authors propose the use of back-propagation to produce a fuzzy controller. In this case two kinds of neural networks are trained: the first kind uses simple numerical data to obtain the membership functions, and the second kind is trained with 0s and 1s to obtain the fuzzy rules. The results show that it is possible to obtain a fuzzy controller without too much data to train the nets. Computer simulations were carried out. The controller was used to control the position of a DC motor. The results show a fast response of the motor without overshoot.>