Neural fuzzy controller to control the angle of heel and the course of the unmanned aerial vehicle

The article is devoted to the method of constructing, training and testing a fuzzy neural network (FNN) which is used as regulator to control the angle of heel and the course of the unmanned aerial vehicle (UAV). For the implementation of neuro-fuzzy controller applies the adaptive algorithm of self-organization network. For constructing the architecture of the FNN used the source data of a special type, the structure of which is fully described in the article and the Gaussian membership functions. The results of experimental studies on the selection of the parameters of the membership function: degree of curvature, Euclidean distance and using designed controller are given.