Real-Time Model Predictive Control For Stability Of Four-Wheel Independent Drive Electric Vehicle

The four-wheel independent drive electric vehicle can obtain higher handling stability through completely independent four-wheel torque control.However,the increase of independent controllable degrees of freedom makes the control system more complex.The complex vehicle stability control system with strong coupling and nonlinear characteristics challenges the real-time control optimization,especially for the model predictive control(MPC)method which has been widely used in recent years.In this paper,an integrated controller of active front wheel steering(AFS)and direct yaw moment(DYC)is designed for four-wheel independent drive electric vehicle based on MPC theory.In order to shorten the computing time of MPC,on the one hand,a convex optimization tool CVXGEN is used to generate MPC algorithm based on real-time iterative scheme for dynamic optimization;on the other hand,an improved MPC unconstrained analytic algorithm is designed to improve the degree of system input and output satisfying the constraints.A MPC constrained optimization algorithm based on quadratic programming(QP)is given to verify the feasibility and real-time performance of the proposed methods.The main research contents are as followsFirstly,a hierarchical control structure is designed by taking the integrated controller of active front wheel steering and direct yaw moment as an example: in the upper controller,the handling stability is taken as the optimization objective,the model predictive control method is used to calculate the expected additional front wheel angle and yaw moment according to the driver input intention and vehicle driving state;in the lower controller the attached ellipse is considered as the constraint condition,the weighted least square(WLS)method is utilized to optimize the distribution of additional yaw moment to each wheel.Through simulation analysis and comparison,the controller can effectively make the vehicle track the ideal state,so as to improve the vehicle handling stability.Secondly,for the sake of reducing the calculation time of MPC,based on the analysis of actuators saturation constraints and control system state constraints regions of four wheel independent drive electric vehicle,an improved MPC unconstrained analytic algorithm is proposed,which can obtain the effective value of the optimal solution of the online optimization problem by introducing the fuzzy constraint region to restrict the optimal solution before solution violates the constraint.On the basis of mastering the realtime convex optimization tool CVXGEN,combining it with MPC optimization problem of integrated controller,a high-speed solution code is generated.The real-time performance of the proposed algorithm is demonstrated by comparing with other algorithms,.Finally,based on the experimental car being developed by the author’s research laboratory,a remote control control system of four-wheel independent drive electric vehicle is designed and built,including PC as the upper controller,chassis communication controller,in-wheel motor drive controller,etc.By debugging and preliminary operation test,the stability control of four-wheel independent drive electric vehicle is realized.The simulation and preliminary experimental results show that the designed hierarchical controller and the proposed algorithm are basically feasible,which can effectively maintain the stability of the vehicle under different conditions and meet the real-time requirements.