Research on Lateral Stability Control Strategy for Distributed Drive Electric Vehicles Considering Driving Style

To improve the handling stability of distributed drive electric vehicle (DDEV), a lateral stability control strategy considering driving style is proposed. Firstly, a driving style recognition model based on support vector machines (SVM) was constructed to identify driving styles in different lateral motion scenarios. Then, a layered architecture is used to design a driver/Active Front Steering (AFS)/Direct Yaw Control (DYC) stability coordination control strategy. The upper layer is based on the phase plane of the sideslip angle and extension control theory, dividing the vehicle operating range into classical, extension, and non domains. In the middle level, three control strategies are designed for different work domains. In the classical domain, a driver/AFS coordinated control strategy based on non-cooperative Nash games is constructed, and the weight coefficients in the AFS system cost function are modified based on driving characteristics; Construct a coordinated control strategy for AFS/DYC systems based on cooperative Pareto games in the extension domain; In non-domain, construct a direct yaw torque control strategy based on twin-delayed deep deterministic policy gradient (TD3) reinforcement learning. The lower layer aims to minimize tire utilization and optimizes the allocation of additional yaw moment. Finally, by establishing a Hardware in the Loop (HIL) experiment platform, the effectiveness of the proposed control strategy is verified in lane changing scenarios of different styles of drivers. Compared to the DYC stability control based solely on TD3, the maximum yaw rate and standard deviation of the lateral stability control strategy considering driving style reduce by 2.4% and 10.82%, respectively, and the maximum and standard deviation of the sideslip angle reduce by 29.29% and 39.24%, respectively, thereby improving the vehicle's handling stability.