Autonomous vehicle following by a fuzzy logic controller

This paper presents a fuzzy logic for performing autonomous vehicle following which is defined as a vehicle controlling its own steering and speed while following a lead vehicle. This controller functions by first converting the range and heading angle observation into fuzzy variables, by executing fuzzy rules generated by a self-organizing neural network, and then by defuzzifying the resultant fuzzy steering-wheel angle and speed variables. In contrast to model-based controller, the developed fuzzy controller constitutes model-free approach. A noise analysis has been carried out to evaluate the performance of such a controller in the presence of observation longitudinal vehicle. The performance of this controller is compared to that of a previously developed model-free neural network controller