ADAPTIVE NEURAL CONTROLLER FOR VEHICLE LATERAL POSITION BASED ON GENETIC ALGORITHM

Abstract

In this paper, neural controller is proposed to control on the vehicle lateral position. The structure of the controller used consists of modified Elman recurrent neural networks that learned on-line by using genetic algorithm teachings. Using of both right and left braking forces simultaneously has automatically kept the vehicle in the correct path when the vehicle is going through an emergency situations. Therefore, it is used as feedback neural controller that is learned on-line in order to control the transient state output of the system by minimizing the error between the actual output of the system and the model reference output. The evolutionary techniques based on this algorithm are employed for the model-reference adaptive control scheme for this system. The variations of the vehicle parameters are addressed in controller. Computer simulation results, using a nonlinear vehicle model are included, and show the feasibility of the proposed controller.