改进强跟踪滤波算法及其在汽车状态估计中的应用  

Improved Strong Track Filter and Its Application to Vehicle State Estimation

作　　者：周聪[1] 肖建[1] 

机构地区：[1]西南交通大学电气工程学院,成都610031

出　　处：《自动化学报》2012年第9期1520-1527,共8页Acta Automatica Sinica

基　　金：国家自然科学基金(51177137;61134001);中央高校基本科研业务费专项资金(SWJTU11CX034)资助~~

摘　　要：准确实时地获取汽车行驶过程中的状态变量,对汽车底盘控制有着重要的意义,而这些关键状态往往难以直接测量或者成本较高.结合纵向、侧向和横摆三自由度非线性汽车模型,将改进强跟踪滤波(Improved strong track filter,ISTF)算法应用到汽车的状态估计中,并改进了算法的稳定性.与扩展卡尔曼滤波(Extended Kalman filter,EKF)算法进行比较分析.通过Carsim和Matlab/Simulink联合仿真和实车双移线实验验证算法,结果表明,该算法在估计精度、跟踪速度、抑制噪声等方面均优于扩展卡尔曼滤波算法,满足汽车状态估计器的软件性能要求.Accurate and real time acquirement of vehicle state variables in running is of great significance to vehicle chassis control. However, these key state variables are not easy to measure directly or cheaply. An improved strong track filter （ISTF） which is much more stable is introduced in this paper. By using a nonlinear 3 degree-of-freedom vehicle model including longitudinal motion, lateral motion, and yaw motion, a state estimation algorithm was established and applied to vehicle state estimation. Comparison was made between extended Kalman filter （EKF） and ISTF. A double lane change test was carried out on Carsim and Matlab/Simulink co-simulation as well as on a real vehicle. The results showed that ISTF is better than EKF in the estimating accuracy, tracking speed, and restraining noise. It was proved that ISTF can satisfy the requirements of vehicle state estimation.

关 键 词：汽车动力学 状态估计 卡尔曼滤波器 强跟踪滤波器 

分 类 号：TN713[电子电信—电路与系统] U467[机械工程—车辆工程]