A Study On Triaxial Force Estimation Of Intelligent Tire Based On Sensitivity Analysis And Radial Basis Neural Network

With the continuous improvement of information technology and the level of the automobile industry,the improvement of automobile safety performance by automobile informatization and intelligence has attracted more and more attention from the industry.As the only automotive component that contacts the road,tire transmits the forces and moments required for vehicle acceleration,deceleration and steering.They are critical to the safety and steering stability of the vehicle,but compared to a variety of vehicles chassis control system,currently tire is still regarded as a passive rubber component,and its role in the field of vehicle safety is not fully reflected.Intelligent tire use sensor technology to monitor tire dynamic information in real time and estimate vehicle dynamics-related information such as tire force and road adhesion characteristics,which has important theoretical significance and engineering application value for improving vehicle safety control.In this paper,the finite element model of 205/55R16 radial tire is established based on ABAQUS finite element simulation software,and the effectiveness of the finite element model is verified by static ground contact test and tire stiffness test.In order to explore the location and type of sensors placed on the inner liner when estimating the tire force,a global sensitivity analysis method based on variance was used to analyze the contribution of the acceleration,displacement and strain of triaxial force of the tire on different areas of the inner liner during tire rolling.The results show that the sensitive response area of vertical force is crown,shoulder and sidewall,and the sensitive response includes three types of response components of acceleration,displacement and strain;the common sensitive response area of longitudinal force and vertical force is crown and sidewall.The circumferential displacement of the crown and sidewall and the radial acceleration of the sidewall and rim area can be used to estimate the longitudinal force of the tire;the sensitive response area of the lateral force is the sidewall,according to the lateral acceleration and lateral displacement response signal can be used to estimate the lateral force.The sensitivity analysis results provide theoretical guidance for the layout of smart tire sensors.In order to obtain the information related to the tire force from the raw data of the sensor signal,the characteristic values of the sensitive signal of the tire force are summarized,and the correlation between the typical characteristics and the corresponding tire force is analyzed.The triaxial force prediction model based on radial basis neural network.The differences in the input structure of the neural network due to the number,type and installation location of the sensors are studied,which in turn affects the estimation accuracy of the prediction model.Therefore,from the perspective of theory and practical application,the sensor layout scheme for the triaxial force prediction of smart tires is proposed.