Research On Intelligent Vehicle Sideslip Angle Estimation And INS/GPS Information Fusion Method

Intelligent driving technology is a hot issue in the automotive industry.Maintaining vehicle stability and vehicle pose resolution and precise positioning during autonomous driving is a key issue.Accurately obtaining the key state parameter information of the sideslip angle can more accurately estimate and predict the driving trajectory and actual driving state of the intelligent vehicle in the future,and the real-time accurate sideslip angle information is basis of the stability control of the intelligent vehicle.However,due to technical or cost reasons of respective sensors,the sideslip angle is difficult to measure directly,so in order to obtain a relatively ideal control,need to be accurate and effective estimation.Vehicle position calculation and navigation positioning are another important part of intelligent driving,and it is an important prerequisite for realizing vehicle path tracking.For the problem of GPS signal loss and INS measurement error accumulation caused by occlusion of GNSS signals in complex environments,combined navigation technology is used to reduce costs.Improve the accuracy of vehicle navigation and positioning.This paper focuses on the problem of the method of estimating the sideslip angle of the intelligent vehicle and the method of integrating the navigation information,including the following two aspects:1)Estimation of the sideslip angle of the intelligently driven vehicle.Firstly,4-DOF vehicle estimation model is established,and the magic tire formula is quoted to ensure the accuracy of the model in the nonlinear region.This paper uses the Cubature Kalman filter(CKF)algorithm to estimate the vehicle sideslip angle parameters as this method is relatively simple and the adjustment parameters are few.Meanwhile,no linearization processing is needed,and the filtering effect of strong nonlinear system is better.At the same time,the Extended Kalman filter(EKF)is used to estimate the sideslip angle,and the advantages and disadvantages of the two algorithms in estimating the sideslip angle are compared.In this paper,the effectiveness of the CKF algorithm is verified by the differential GPS real vehicle test.The experimental results show that the estimated centroid angle information of the CKF algorithm is in close proximity to that obtained by the differential GPS system.The algorithm is more precise than the results obtained by the EKF algorithm.2)Integrated navigation based on Improved Sage-Husa Adaptive Kalman Filtering(SHAKF).Firstly,the paper studies the basic composition and working principle of Inertial Navigation System(INS)and Global Positioning System(GPS).Secondly,the INS error model and the INS/GPS integrated navigation system model are established.The difference between the speed information and the position and speed information of the GPS output is used as the measurement information;the improved Sage-Husa Adaptive Kalman Filter(SHAKF)algorithm proposed in this paper is used to adaptively adjust the statistical characteristics of system noise,which improves the robustness of the system and improves the accuracy of navigation positioning.Finally,the comparison of the actual path trajectory obtained by the proposed algorithm with the traditional SHAKF algorithm verifies the effectiveness.