Research On Active Four-wheel Steering Control And Path Following Of 4WID Electric Vehicle Based On Multi-Agent Framework

The four-wheel independent drive electric vehicles(4WID EVs)utilize in-wheel motors to drive the wheels,which reduces the mechanical transmission device.The torque of each wheel can be controlled independently so that different active control strategies can be realized.The 4WID EVs provide an ideal platform for active safety control and self-driving technology.Meanwhile,as one of the most significant ways to achieve the active safety control,the four-wheel steering(4WS)technology can improve the maneuverability of vehicles at low velocity and the handling stability of vehicles at high velocity.This thesis proposes a dynamic control framework.An active four-wheel steering controller is designed based on differential game.An active four-wheel steering path following control strategy is developed based on model predictive control.The speed tracking experiment is carried out based on the 4WID EV experimental platform.The main research contents are as follows:(1)A dynamic control framework is proposed for 4WID EVs to provide foundation for the design of the subsequent control systems.The 4WID-4WS EV mathematical model and the driver models are built and analyzed.The detailed definition of vehicle system characteristic parameters is confirmed in CarSim.The in-wheel motors torque response model,the magic formula tire model,the single-point preview driver model for direction control and the driver model for speed control based on fuzzy PID control method are designed in Simulink.The co-simulation using CarSim and Matlab/Simulink is carried out,which verifies the effectiveness and the rationality of the models and provides the model foundation for the design of the control systems.(2)On the basis of analyzing the interaction between the front-wheel steering angle and the rear-wheel steering angle and the conflict problem of the control targets,an active 4WS controller is designed by the combination of the feedforward control and the feedback control.The feedback control of the compensated front-wheel steering angle and the feedback control of the rear-wheel steering angle are defined as two Agents participating in differential game.The non-cooperative feedback Nash equilibrium solution is obtained by solving coupled algebraic Riccati equations.Simulations in the co-simulation environment including CarSim and Matlab/Simulink are carried out and the results are compared with the LQR controller and the proportional 4WS controller,which verifies the effectiveness of the proposed controller.(3)An active 4WS path following controller is developed based on model predictive control.Under the premise of considering the maximum longitudinal speed in simulation conditions,the simulations are carried out to verify the effectiveness of the designed controller in the co-simulation environment including CarSim and Matlab/Simulink.The results are compared and analyzed with the front-wheel steering path following controller.(4)The field experiment is carried out based on the 4WID EV experimental platform.A longitudinal speed tracking controller is designed based on the PID control algorithm.The field experiment is conducted using MicroAutoBox and ControlDesk and the tracking of the desired longitudinal speed is successfully realized.