Control of Four-Wheel-Steering Vehicle Using GA Fuzzv Neural Network

摘要

In order to improve the handling and stability of four-wheel-steering (4WS) vehicle, a new 4WS intelligent control system with genetic algorithm (GA) fuzzy neural network (FNN) was put forward. According to the tire cubic formula, a vehicle nonlinear dynamics model was built. Then a vehicle model based on back-propagation (BP) network was identified from the vehicle dynamics. Next a fuzzy neural network controller was designed. Speed, steering angle of front wheel and lateral acceleration were taken as its input variables and steering angle of rear wheel was taken as its output variable. At last GA was used to optimize the fuzzy neural network controller. The results of computer simulation demonstrate that the 4WS intelligent control system with GA fuzzy neural network can markedly reduce the side slip angle and the yaw rate compared with linear 4WS control law and 2WS control law, the entire control system is robust, and the optimization by GA improves the control performance of controller and the design efficiency.