Study On Methods Of States Estimation And Stability Control Research For Wheel Drive Electric Vehicle

With the increasing severe challenges of the environment, energy for vehicles, electric vehicles are becoming the focus of research and development of the global automotive industry. Due to the flexible dynamic layout, the traditional mechanical connection on the car is eliminated, which lead to simplify the vehicle structure greatly and increase in degrees of freedom of the driving wheels. Also, with its unique advantages on structure and control, it is easier to implement the integrated chassis control system. However, it also brings problems like stability control to wheel drive vehicles. During stability and safety control, most of these control systems need to know some key vehicle state information, such as side-slip angle, yaw rate and the longitudinal speed, etc. It can be difficult to acquire all the needed information by the on-board sensors alone. In order to achieve stable control for wheel drive electric vehicle and solve the problem of difficult acquiring of needed state information due to measuring difficulty and cost restriction, this paper takes wheel drive electric car as the research object, and conducts a systematic study on state estimation and stability control in its driving process. The main work is listed as follows:① Build a wheel drive electric vehicle simulation model. Use a certain electric car as platform and develop a wheel drive system; adopt the virtual simulation software MATLAB/Simulink to create a complete simulation model for wheel drive electric car.② Key state estimation for vehicle chassis integrated control. Accurate and real time acquiring of the driving state information of the vehicle is prerequisite to the stability and active safety control. Use the Extended Kalman Filtering algorithm to estimate the states of the wheel drive electrical vehicle based on 3-dof vehicle model. Through simulation experiments, choose serpentine conditions to verify the accuracy of the estimation algorithm.③ Estimation of tire force and road adhesion coefficient. On the basis of the original state estimation method, use the Extended Kalman Filtering algorithm to estimate tire force and road adhesion coefficient based on the second-order Gaussian Markov process and improved the "magic formula" tire model. Through simulation experiments, choose different road conditions to verify the accuracy of the estimation algorithm.④ Research of stability control. Designed a sliding mode controller to calculate the required yawing moment based on the state estimation results. Studied the weighted least squares driving force distribution method which minimize the tire-load rate. Distribute longitudinal force reasonably under constraint conditions of the road adhesion conditions and maximum motor torque, and verify it with simulation.