UKF滤波方法及其在车辆导航状态估计中的应用(英文)  

Application of Unscented Kalman Filter in State Estimation for Land Vehicle Navigation System

作　　者：张传斌[1] 田蔚风[1] 金志华[1] 

机构地区：[1]上海交通大学仪器工程系,上海200030

出　　处：《系统仿真学报》2005年第6期1456-1458,共3页Journal of System Simulation

摘　　要：在车载导航系统中,通常采用EKF作为状态估计方法提高导航的精度。由于EKF进行非线性估计存在一些缺陷,因此将其用于导航系统的非线性估计时,存在估计误差,从而影响导航系统的精度。为了获得更高的导航精度,将一种新的滤波方法—UKF方法用于车载导航系统的状态估计中。对一个车载DR/GPS组合系统,将EKF和UKF方法分别进行了滤波仿真。仿真结果表明:在车载导航状态估计中,UKF方法优于EKF方法。In a land vehicle navigation system, generally the Extended Kalman Filter (EKF) is as a state estimation method to improve the accuracy of navigation. However, as defects of the EKF in nonlinear estimation, there exists estimated error, which affects the accuracy of the navigation system, when it is adopted in nonlinear estimation of a navigation system. In order to yield the higher accuracy of navigation, a novel method-Unscented Kalman Filter (UKF) was employed in state estimation for a land vehicle navigation system. For a land vehicle DR/GPS navigation system, the EKF and UKF are compared through simulation. Simulation results show that the UKF is superior to the EKF in state estimation for a land vehicle navigation system.

关 键 词：滤波方法 UKF 车辆导航 车载导航系统 应用 状态估计方法 非线性估计 EKF 估计误差 导航精度 组合系统 仿真结果 GPS DR 

分 类 号：U463.6[机械工程—车辆工程] TN713[交通运输工程—载运工具运用工程]