Friction Coefficient Estimation For Distributed Drive Electric Vehicle Based On Tire Sensors

Friction coefficient between tires and road plays very important roles in the traditional vehicle control system.Such as adaptive cruise control(ACC) 、anti-lock braking system(ABS)、driving control system(TCS) 、stability control system(ESP) are all based on the friction coefficient to control the vehicle. For the distributed driving electric vehicle, in addition to the control system, the unique driving mode-each wheel as an independent driving wheel, more accurate estimated friction coefficient between wheel and road surface for each driving wheel is required, However, for the current methods applied for estimating the friction coefficient between the tire and road surface there still exist many problems. For vehicle running stably, observer method cannot update the friction coefficient; when under emergency conditions, the output results of vehicle model are not accurate, all these factors will affect the accuracy of the estimated results. Therefore, this article hopes to try new ways of estimation to improve the accuracy of the estimation results.In this article, about the distributed driving electric vehicle friction coefficient estimation problem which based on the measurement inside of tire, the main research contents include the following aspects:First, reviewed the assessment methods for friction coefficient between tire and road surface and analyzed the main advantages 、disadvantages and scope of application of the current methods, and a brief introduction of the study of intelligent tire testing system was given.Secondly, mainly study the research methods for estimating friction coefficient between tire and road surface which based on measurement system inside of the tire. To verify the feasibility of measurement system inside the tire, measurement system and distributed drive electric vehicle test platform was built, and the constant speed test、 accelerated 、circular and other conditions test of real vehicle were carried out. Side force of tire 、Aligning torque was calculated by Matlab to verify the feasibility and accuracy of the measurement system inside the tire.Then, the article made a contrastive analysis of several common tire models, considering the operation speed and accuracy of the model, finally selected the brush model as the tire model of friction coefficient estimation. The friction coefficient is estimated by bringing the tire force information into the model. In order to improve the estimation accuracy, use EKF method and compare the estimated results with brush model estimator to verify the rationality.Again, for distributed driving electric vehicle, the driving torque and speed of the wheel motor can not only be measured accurately, but also can be controlled accurately. Based on the real-time information of the wheel motor, the longitudinal dynamic model of the vehicle is established, and the longitudinal friction coefficient estimator based on the brush tire model can be derived which can effectively and rapidly estimate the longitudinal friction coefficient.Finally, in order to describe the vehicle motion state more accurately and consider the situation of the vehicle’s lateral and longitudinal motion, in addition, the contact motion process between tire and road surface is not just a static friction process. Therefore, this article reviews the dynamic tire friction model, linking to the vehicle lateral motion, longitudinal motion and Lu Gre tire model, and try to integrate the Lu Gre tire model, tire measurement system, the real-time information of wheel motor to improve the accuracy of the estimated friction coefficient between tire and road surface.