The application of ART neural network to image processing for controlling a mobile vehicle

This paper presents experimental results of pattern recognition for controlling a mobile vehicle by using adaptive resonance theory neural network (abbreviated to ART). The aim is to give a practical method to the pattern recognition by ART. ART, a kind of competitive neural network, can self-organize and self-stabilize in response to complex input vectors. New patterns are stored in such a fashion that existing ones are not forgotten or modified. Finally, experimental results and some future considerations are also given