Vehicle traveling state estimation method based on sparse grid integrating Kalman filtering

Abstract

The invention discloses a vehicle traveling state estimation method based on sparse grid integrating Kalman filtering. Through a selected seven-degree-of-freedom vehicle motion module, on the basis ofan integrating Kalman filtering method combining with a sparse grid theory, multidimensional integration points are selected, time upgrading and measurement upgrading are conducted, and the longitudinal vehicle speed, the lateral vehicle speed and the side slip angle of a vehicle are estimated. The vehicle traveling state parameter estimation method has the characteristic of high precision, and the real-time performance of state parameter estimation can be effectively improved while the state estimation precision is improved.

Description

technical field [0001] The invention belongs to the field of system state estimation, in particular to a method for estimating vehicle running state based on sparse grid quadrature Kalman filter. Background technique [0002] With the in-depth development of vehicle active safety technology, accurate acquisition of the vehicle's own driving state is becoming more and more important. Due to the limitation of sensor technology and vehicle cost, vehicle driving state estimation based on model prediction has become one of the research hotspots. [0003] Currently, vehicle state estimation methods based on model prediction mainly include Extended Kalman Filter (EKF), Unscented Kalman Filter (UKF), Volumetric Kalman Filter (CKF), Particle Filter ( Particle Filter, PF), etc., in which the extended Kalman filter converts the nonlinear system into a linear system by Taylor series expansion, and the error is large when the system nonlinearity is strong; the unscented Kalman filter us...

AI Technical Summary

Problems solved by technology

[0003] Currently, vehicle state estimation methods based on model prediction mainly include Extended Kalman Filter (EKF), Unscented Kalman Filter (UKF), Volumetric Kalman Filter (CKF), Particle Filter ( Particle Filter, PF), etc., in which the extended Kalman filter converts the nonlinear system into a linear system by Taylor series expansion, and the error is large when the system nonlinearity is strong; the unscented Kalman filter uses the Sigma point selection to perform UT transformation avoids the approximation of the nonlinear system, and has high precision, but the calculation amount is large; the precision of the volumetric Kalman filter is equivalent to that of the unscented Kalman filter, but fewer sampling points are required; the precision of the particle filter is high, However, there are problems such as exhaustion of particle samples and poor real-time performance

[0004] Quadrature Kalman filter (Quadrature Filter, QF) is a kind of nonlinear filter, its estimation accuracy is higher than EKF, UKF, CKF, but its real-time performance is poor, because its calculation amount will increase exponentially with the increase of state dimension rise

Images