基于Doppler量测的CS-EKF机动目标跟踪平行滤波算法  

Parallel Filtering Algorithm for CS-EKF Maneuvering Target Tracking with Doppler Measurement

作　　者：张喜涛[1] 张安清[1] 

机构地区：[1]海军大连舰艇学院,辽宁大连116018

出　　处：《现代防御技术》2015年第3期119-123,145,共6页Modern Defence Technology

摘　　要：为提高非线性机动目标跟踪精度,在基于"当前"统计模型(CSM)的扩展卡尔曼滤波(CS-EKF)算法的基础上,提出一种基于多普勒径向速度量测和三维平行滤波的机动目标跟踪算法(CS3D-EKFrv)。该算法通过引入径向速度量测扩充量测矩阵的维数,然后利用扩展卡尔曼滤波(EKF)方法解决量测方程中状态向量和量测向量的非线性问题,最后采用"当前"统计模型对目标的三维状态进行平行滤波估计,解决三坐标轴上机动强度不一致的问题。对CS-EKF,CS3D-EKF及CS3D-EKFrv算法的仿真结果和实测数据检验表明,CS3D-EKFrv算法能够有效改善机动目标的跟踪精度。To improve the tracking precision of maneuvering target, based on the extended Kalman filtering algorithm （CS-EKF） on the basis of current statistical model （CSM） , an extended Kalman filte- ring algorithm with the radial velocity measurement and a parallel algorithm for three dimensions of Carte- sian coordinates （CS3D-EKFrv） are proposed. In this algorithm, the dimension of measurement matrix is extended by introducing the radial velocity measurement, and then the non-linearity of the state vector and measurement vector in measurement equations are solved by using the EKF algorithm. Finally the es- timation for target sate of three dimensions can be done by using CSM parallel. The simulation results of CS-EKF, CS-EKF with parallel filtering for three dimensions （CS3D-EKF） and CS3D-EKFrv algorithms show that the CS3D-EKFrv algorithm could effectively improve the tracking precision and the convergence rate of maneuvering target, which is also proved by actual measurements.

关 键 词：目标跟踪 “当前”统计模型 扩展卡尔曼滤波 径向速度 

分 类 号：TN953[电子电信—信号与信息处理] TN713[电子电信—信息与通信工程]