Distributed electric vehicle decoupling control based on GA-BP neural network

Aiming at the coupling interference phenomenon of distributed electric vehicle in longitudinal and lateral motion, a decoupled controller using genetic algorithm optimism BP neural network (GA-BP) is proposed. The top controller is designed as GA-BP neural network decoupling controller, the decoupling linearization system is established based on the principle of neural network inverse system, the neural network is constructed and trained, the weights and thresholds of BP neural network were acquired, and the optimal value is obtained by GA algorithm. However, the lower controller is designed to take the minimum tire loading rate as the objective function, and the quadratic programming algorithm is adopted for the online optimization of the system. Co-simulation based on Carsim and MATLAB/Simulink is carried out to verify the effectiveness of the control strategy. The results show that the proposed GA-BP controller has good decoupling characteristics and achieves the effect of independent controllability of the vehicle longitudinal and lateral systems, small controllable range of the side-slip angle, and improved tracking accuracy of the yaw rate, which improves the mobility and driving stability of the vehicle.