| In recent years,real-time vehicle data provided by a CAN bus-based automotive information platform has provided conditions for real-time estimation of vehicle mass and road slope.However,due to the coupling relationship between vehicle mass and road slope and the complexity of driving conditions,the relationship between vehicle driving data,vehicle mass and road slope is difficult to determine.There are no specific solutions for estimating vehicle mass and road slope under consideration of the coupling relationship between parameters and the influence of complex conditions.Therefore,in order to solve these problems,it is important to study more stable,accurate and robust methods for jointly estimating vehicle quality and road gradient.It is of great practical significance for realizing reasonable vehicle shift control,economic vehicle speed control,and vehicle stability control.It is also the key to intelligent driving.Based on the Ford Motor Company developed OpenXC devices real-time access real-time vehicle driving state data,for the estimation distortion and filter divergence brought about by the coupling relationship in the estimation,this paper studies the vehicle mass and road slope estimation model combined with the recursive least-squares method with forgetting factor and the extended Kalman filter algorithm.In order to solve the problems of vehicle mass and road slope estimation from the mechanism modeling under the complex braking and turning conditions,an improved approach based on state preservation and segment estimation was proposed which improves the robustness of vehicle mass and road slope estimation methods in actual driving and expands the scope of application of the method.Research work of this paper includes the following three aspects:Firstly,based on the car driving data obtained by the OpenXC and combining the longitudinal dynamics of the vehicle,a method for estimating vehicle mass and road slope based on the recursive least squares method with forgetting factor and the extended Kalman filter is established.The contrastive experiments of CarSim simulation platform are used to compare the effect of the algorithm and the characteristics of the parameters,through the analysis of the coupling mechanism of the vehicle mass and the road slope estimation,the experimental error transmission characteristics under this coupling relationship are obtained.Secondly,based on the coupling mechanism and error transfer characteristics ofvehicle mass and road slope estimation,a joint RLS-EKF estimation method was established.By analyzing the effect of different estimation steps on the estimation of vehicle mass and road slope,a reasonable estimation step is obtained.Based on this,combined with the parameters of vehicle mass and road slope,an improved RLS-EKF joint estimation method for data anomalies in the estimation process was established.Experiments compare the estimated performance of RLS,EKF,and RLS-EKF.The results show that RLS-EKF has better stability and accuracy.Thirdly,aiming at the difficulty of analyzing the braking and turning conditions from the longitudinal dynamics of the vehicle,by analyzing the characteristics of OpenXC data under braking and turning conditions and the influence of different brake and turn data proportions on the joint estimation effect,an improved method of state preservation is proposed for single estimation,and based on this,the whole estimation system is further proposed.Real vehicle experiments show that the improved joint estimation algorithm adapts to the traffic environment of a complex urban road and overcomes the estimation bias of single RLS-EKF algorithm in braking and turning. |
PDF Full Text Request