Path Following Control For Four-wheel Independently Driven Electric Vehicles Based On Differential Steering

The developmemt of unmanned vehicles requires that the platforms have dependable security,efficiency and flexibility.Therefore,the emerging four-wheel independently driven(FWID)electric vehicle(EV)is a promising architecture,which can provide independent actuation,can bring flexible actuation,fast and precise torque and speed responses.The FWID unmanned vehicle can combine the advantages of independent actuation and unmanned vehicle,and is a promising solution to achieve greater road safety.The differential steering mechanism can be utilized as a fault-tolerant mechanism to deal with the failure of Steer-By-Wire system.When the steering motor fails and the effective steering of the vehicle cannot be realized,the differential steering mechanism is intervened to ensure the safe driving of the vehicle.This paper applies this kind steering mechanism to the unmanned electric vehicle,and designs the control strategy to further study the path tracking problem of the unmanned vehicle.The main research work of this paper is as follows:First,the basic principle of the differential steering mechanism is expounded,and the dynamics of the differential steering system is modeled and analyzed.Considering the nonlinear brush tyre model,the vehicle dynamic models based on the differential steering and the active front wheel steering are built.The path following model is established with the lateral error and the direction angle error as the state quantity.The objective of the path following is to design a controller to globally stabilize lateral error and heading error and asymptotically converge them to zero.Second,the desired value of the yaw rate is redefined based on the path tracking error,which translates the path following problem into the vehicle stability problem.Tracking the desired yaw rate and lateral velocity can ensure the path following task as well as maintaining the stability of the vehicle.In this paper,the upper path following controller is designed based on the linear time-varying model predictive control,then the lower controller is built to distribute the differential torque to the driving wheels.A state observer based on the kinematic model is designed to estimate the vehicle longitudinal and lateral velocities and sideslip angle,which are difficult to be accurately measured only through the low cost sensors.The estimation accuracy of the observer is verified with simulation cases at a high speed on high adhesion road and at a high speed on high adhesion road condition.Finally,in the CarSim-Simulink joint simulation platform,the effectiveness of the path tracking controller and the feasibility of the differential steering are proved.Third,a vehicle test is carried out based on the four-wheel independently driven electric vehicle test platform,which includes open-loop tests and close-loop tests.In open-loop tests,two kinds of vehicle tests are carried out,which are the performance test of the test vehicle and the open-loop verification of the differential steering mechanism.The former analyzes the acceleration,deceleration and steering characteristics,which paves the way for the feasibility verification of the differential steering mechanism.The latter verifies that the four-wheel independently driven electric vehicle can realize the differential steering function.Finally,the close-loop experiments are investigated,and the PID control algorithm is used to track the desired yaw rate.It is proved that the electric test vehicle based on the differential steering can track the constant and variable yaw rates.The open-loop and close-loop tests of the real vehicle achieve satisfactory results,which prove the feasibility of the differential steering mechanism.