关键词: 汽车动力学；双重扩展自适应卡尔曼滤波；状态和参数估计；虚拟试验

Abstract:

A critical component of vehicle dynamic control systems is the accurate and real time knowledge of vehicle key states when running on road. A new vehicle states estimator was proposed. First the nonlinear vehicle dynamics system was established in which contained inaccurate model parameters and unknown time varying noise. Then a dual extended adaptive Kalman filter (DEAKF) algorithm was proposed. In the algorithm two Kalman filters run in parallel, state estimation and parameter estimation update each other. In order to be robust to unknown time varying noise, the noise statistical estimator was inserted between state correction and parameter correction. The results of virtual experiment based on ADAMS and real vehicle experiment demonstrate that the DEAKF algorithm has higher state estimation accuracy than that of EKF and DEKF, also has good capability to revise model parameters. The conclusions can provide theoretic direction for design of estimator in vehicle stability control system.

Key words: vehicle dynamics, dual extended adaptive Kalman filter, state and parameter estimation, virtual experiment