Research On Improvement Of Four-Wheel Drive Electric Vehicle Handling And Human-Vehicle Closed-Loop System

Because of its green and clean characteristics,electric vehicles will become the primary tool for future travel.At the same time,four-wheel drive electric vehicle is currently a research hotspot in the direction of dynamics control due to their controllable degrees of freedom and fast response of the executing components.This paper mainly researches from two aspects:One is the lateral dynamics control of the chassis.Based on the characteristics of four-wheel independent drive,a torque vector control strategy that can improve the vehicle's handling performance can be developed.The other one is to explore the relationship between control quality and control characteristic parameters.Based on the human-vehicle closed-loop system,a large number of variable-parameter experiments are performed by using a driver model to analyze the relationship between the handling characteristic parameters and the handling evaluation.First,the development history of electric vehicles and its necessity are reviewed.The previous four-wheel-drive dynamic control methods and the closed-loop evaluation system based on human-vehicle are reviewed.According to the simulation requirements and the actual vehicle test requirements,a higher-precision Carsim and Matlab/Simulink joint simulation dynamic model is established,which mainly includes the adaptation of Carsim vehicle models,the construction of external driver models and motor models.And then,the whole model was validated based on real vehicle data.After that,the development and simulation verification of torque vector control strategy for steering handling were carried out.Based on the two-degree-of-freedom vehicle model,the basic mechanism of maneuverability control and the control principle of direct yaw moment control were discussed,and a hierarchical control structure was proposed.Among them,as the upper yaw moment decision-making layer,three control methods of first-order sliding mode,second-order sliding mode,and adaptive second-order sliding mode based on RBF are gradually derived to determine the additional yaw moment.The first-order sliding mode control needs side-force information,but the RBF-based adaptive second-order sliding mode control can adaptively estimate external interference and does not require side-force information which can improved the algorithm adaptability and robustness.The lower layer designs a torque distribution strategy based on minimizing tire load.The quadratic programming method is used to minimize the solution,and then the required torque of the four wheels is obtained.Then,angular pulse simulation test is selected to verify the feasibility of following the desired handling characteristics.The high-adhesion and low-adhesion double-line changing contrast simulations are selected to verify the effectiveness of the algorithm.Then,based on the human-vehicle closed-loop system,the relationship between handling characteristics and handling quality was studied.A lateral driver model based on the preview theory was established.Based on the research on the fixed preview time model,an adaptive correction method of the preview time was proposed.After that,a steering maneuverability evaluation scheme including tracking maneuverability,driver's control burden and safety is proposed.Finally,using the ability that torque vector control can change steering characteristics,a large number of human-vehicle closed-loop tests with variable target parameters were carried out.Under the evaluation scheme system in this paper,The relationship between natural frequency ?_n,damping ratio?and handling quality are summarized.Finally,a four-wheel drive electric vehicle was used for real vehicle testing.First,VCU-in-loop tests are carried out to verify the stability of its basic interface functions and algorithm operation.After modifing and debugging basic control of electric vehicles,the vehicle of which genuine VCU were replacd can achieve basic driving functions.And then,the actual vehicle test on the high adhesion road and low adhesion road was carried out.The double-lane-change test and snakering test are selected for the comparison test of the algorithm opening and closing.The results show that the torque vector control strategy designed in this paper can effectively reduce the driver burden,improve the vehicle handling,and increase the maximum passing speed.