Four Wheel Steering Vehicles Stability Control Based on Adaptive Radial Basis Function Neural Network

Due to the nonlinear and strong coupling characteristics of four-wheel steering vehicles, there are poor control accuracy and robustness in traditional control methods. This paper proposes a hierarchical controller of lateral stability based on mechanical-differential combined distributed drive four-wheel steering vehicle. The upper controller trains the RBF neural network offline based on the ideal sideslip angle, the parameters adaptively adjusted online according to the error and calculates the ideal rear wheel angle by using the front wheel angle in real time. In addition, a lower controller is established for the differential torque distribution of the hub motor on both sides with the PI controller. Finally, the performance is evaluated via MATLAB/Simulink. The simulation results prove that the four-wheel steering strategy based on adaptive RBF neural network controller (ADP-RBF) can effectively improve steering mobility at low speed and the lateral stability at high speed of the vehicle.