Unscented卡尔曼滤波对目标位置预测  

Target position prediction of Unscented Kalman filter

作　　者：林瑞阳 杨东升[1] 邱锋[2] 

机构地区：[1]西北工业大学航天学院 [2]中国人民解放军96361部队

出　　处：《现代电子技术》2014年第1期34-37,共4页Modern Electronics Technique

摘　　要：采用一种针对目标位置预测只能测量角度信息的卡尔曼滤波算法,实现对目标的位置、速度和加速度的估计。由于是纯方位目标运动分析,所以一般的线性滤波方法不能使用,主要使用UKF滤波算法,并给出了具体步骤。通过仿真运算与以前的方法进行比较,发现该算法实现方便,并在滤波精度、稳定性和收敛时间等方面有了很大提高。A Kalman filter algorithms is provided to estimate the position, velocity and acceleration of the target, which could only measure the angle information of the target location prediction. Since it is the bearings-only target motion analysis, the general linear filtering methods can not apply. UKF filtering algorithm is mainly used in this paper, and the specific steps are given. Through the simulation calculation and comparison with the previous methods, it is found that this algorithm is easy to be implemented, and has great improvement in filtering accuracy, stability and convergence time.

关 键 词：纯方位目标运动 非线性滤波 UKF EKF 

分 类 号：TN953-34[电子电信—信号与信息处理]