Research On Tire-road Friction Coefficient Estimation Algorithm Of Distributed Drive Electric Vehicle

In recent years,distributed drive electric vehicles(DDEV)get more attention from more people with the advantage of independent wheel control,and people have put forward higher requirements for their active safety and stability control.The premise of active safety control is to obtain the tire-road friction coefficient quickly and accurately for the vehicle active safety control system to make corresponding control decisions for different driving conditions and road surface conditions to ensure the vehicle stable driving,therefore,the accurate estimation of the tire-road friction coefficienthas become the problem that needs to be solved in the field of vehicle dynamics and control.Most of the tire-road friction coefficient estimation methods require the wheel angular speed sensor signals,but the sensors have a limited application range and insufficient reliability in extreme cases.In order to reduce the use of sensors,this paper proposes an algorithm for estimating the tire-road friction coefficient of distributed drive electric vehicles using Permanent Magnet Synchronous Motor(PMSM)sensorless control.The rotor speed obtained by the PMSM sensorless control system is used instead of the wheel angular speed sensor signal to estimate the the tire-road friction coefficient.The research done in this paper for this method is as follows:(1)The research background of tire-road friction coefficient estimation and the current research status of distributed drive electric vehicles are introduced,and the experimental-based and model-based tire-road friction coefficient estimation methods are analyzed and compared,and the model-based tire-road friction coefficient estimation method is selected as the research direction.(2)A 3 degrees of freedom vehicle dynamics model and a Dugoff tire model were constructed,and the three kinds of tire models(the magic tire model,the steady-state exponential unified model and the Dugoff tire model)were analyzed and compared,and the Dugoff tire model was finally selected and simply deformed for subsequent algorithm design.The internal settings of Carsim software are introduced,two motor mathematical models for speed control and speed estimation are determined,and the DDEV complete vehicle model is built jointly using Carsim/Simulink,and its validity is verified.(3)The cubature Kalman filter(CKF)is orthogonally transformed and the strong tracking filter(STF)principle is introduced as an asymptotic factor in the calculation of the covariance array to derive a strong tracking cubature Kalman filter(STCKF)based on orthogonal transformation.The strong tracking cubature Kalman filter based on orthogonal transformation(T-STCKF)is derived and applied to the subsequent rotor speed estimation and tire-road friction coefficient estimation.(4)The HSTCKF-based TRFC estimation algorithm is proposed.Firstly,the vehicle state parameters are estimated using the HSTCKF algorithm,and based on this,the TRFC is estimated again using the HSTCKF algorithm,and the effectiveness of the algorithm is verified through simulation and experiment.(5)The PMSM sensorless control system with adaptive exponential reaching law for sliding mode speed control and T-STCKF algorithm is constructed.A tireroad friction coefficient estimation algorithm using PMSM sensorless control is proposed by substituting the estimated rotor speed instead of the wheel speed sensor signal into the tire-road friction coefficient estimation algorithm.Simulation and experimental comparison analysis of the T-STCKF algorithm based on the wheel angular speed sensor signal,the T-STCKF algorithm using PMSM sensorless control,and the STCKF algorithm using PMSM sensorless control for tire-road friction coefficient estimation are performed to illustrate the effectiveness of the proposed algorithm.