基于“当前”统计模型的后向迭代CKF算法研究  

Backward Iterative Improved Cubature Kalman Filter Based on Current Statistical Model

作　　者：吴博[1] 刘鹏远[1] 李宝晨[1] 

机构地区：[1]军械工程学院,河北石家庄050003

出　　处：《现代防御技术》2016年第3期38-43,56,共7页Modern Defence Technology

摘　　要：针对机动目标的非线性跟踪问题,提出了基于"当前"统计模型的容积卡尔曼滤波(CSCKF)。针对新算法对目标加速度阶跃机动跟踪延迟过大的问题,研究了后向迭代算法在估计精度和动态响应速度上的特点,提出了基于"当前"统计模型的后向迭代容积卡尔曼滤波算法。最后通过仿真实验比较了本文提出的CS-CKF算法和CKF算法的滤波效果。结果表明CS-CKF算法滤波效果优于CKF算法。In order to solve the problem of nonlinear maneuvering target tracking, a new cubature Kalman filter based on ＂current＂ statistical model （CS-CKF） is studied. According to the problem of excessive delay on the step acceleration maneuvering during target tracking, a backward iterative algorithm that amends the last state estimation with the predicted current state is analyzed and applied for improring. Finally, the two kinds of typical maneuvering models are applied to CKF and CS-CKF, and the effects of the two algorithms are compared. The simulation results show that CS-CKF algorithm is better than the CKF algorithm.

关 键 词：机动目标跟踪 非线性滤波 “当前”统计模型 容积卡尔曼滤波 后向迭代算法 动态特性 

分 类 号：TJ765[兵器科学与技术—武器系统与运用工程] TP391.9[自动化与计算机技术—计算机应用技术]