Fuzzy Gaussian Neural Network Controller and Its Application to the Control of a Mobile Robot

This paper describes a new fuzzy neural network (FNN) controller in which a Gaussian function is applied as an activation function, referred to here as a fuzzy Gaussian neural network (FGNN) controller. The learning architecture adopted is specialized so that we can tune the membership functions without using an expert's manipulated data. As an example of application, a tracking control problem for the speed and azimuth of a mobile robot driven by two independent wheels is solved using the FGNN controller. In order to facilitate the application of an FGNN controller to multi-input multi-output systems, we then propose a learning controller consisting of m FGNNs based on independent reasoning and a connection network, where m denotes the order of output of the control object. The effectiveness of the proposed method is illustrated using computer simulation.