On The Stability Control Of Four-wheel-independent-drive Electric Vehicles

With the sharp growth in vehicle ownership, energy crisis and the traffic safety problemincreasingly prominent, the electric vehicle’s development and advanced vehicle safetytechnology caused wide attention from all walks of life. Electric vehicles not only has highutilization efficiency of energy and environment friendly advantages, but because of theelectric vehicles using motor drive system, the torque is easy to accurate measurement, fastresponse and the power can realize distributed control, which can be used to research thedynamics of the new control structure and method and also can significantly improve energysaving and safety control of the vehicle performance.In this paper, a four-wheel-independent-drive electric vehicle as the research object, andfrom two aspects: longitudinal stability control and lateral stability control to research thestability problem of electric vehicles on the road.1. The longitudinal stability mainly prevent the vehicle the skid when driving on thechanged road condition (especially the wet road). In this paper, a sliding mode observer(SMO), based on LuGre tire dynamics model and vehicle longitudinal dynamics model, isproposed. By using of the four independent motor torque and angular velocity of electricvehicles (EV), we can estimation LuGre dynamics model parameters θ of each tire, whichcan indicate the load condition. Then according to the steady-state theta LuGre model, wecan calculate the optimal slip and the maximum friction coefficient of the road, and each tirecan obtain the maximum traction. So we can limit the motor torque to make the electricvehicle drive with maximum acceleration without slipping. Finaly we use the CarSim andMatlab/Simulink simulation experiments to implement the control strategy that the electricvehicle drive in different road conditions with maximum acceleration. Vehicle simulationexperiments can real-time and accurately estimate the corresponding road surface frictioncoefficient of each wheel, achieved good control performance, on different adhesioncoefficient of road surface.2. The lateral stability mainly use direct yawing moment control (DYC) to solveunstability problem of the electric vehicle. At the same time as much as possible to track thedriver’s expectations, improve vehicle handling performance. A method is proposed basedon using the tire lateral force information and the vehicle dynamics model, and use theextended kalman (EKF) to estimate the sideslip Angle of the electric vehicle, in thispaper. At the same time we use the estimated the vehicle sideslip Angle and dynamicmodel of the forgetting factor recursive least squares (FFRLS) to estimate the tire corneringstiffness. By estimating state variables of the vehicle, a lateral stability control system isdesigned. The proposed control system is composed of Upper and lower level control stages.Upper level controller is composed of feedforward and dynamic sliding model (DSM) feedback control algorithm. By controlling the direct yawing momentM z, the vehicle iscontrolled to track the desired transverse angular velocity d, and restrict the vehicleside-slip Angle in a stable range，at the same time. The dcan be obtained according to thevehicle dynamics, which need the given steering Angle of the steering wheel and the vehiclesideslip Angle. According to the restricted conditions (such as road conditions, tire frictioncircle characteristics, etc.), the lower level controller optimize the allocation of each wheeldriving moment. Finally, A CarSim and matlab/Simulink simulation experiment is used toimplement the proposed control strategy, and compared with the classical PI controlalgorithm. In the simulation experiments, the performance of dynamic sliding model (DSM)controller is better than PI controller. And the expectation dobtained from dynamicsmodel proposed in this paper is better reflect the driver’s intentions than the general model.