集中式多传感器无极联合概率数据互联算法

Journal of Systems Engineering and Electronics ›› 2009, Vol. 31 ›› Issue (11): 2602-2606 .

集中式多传感器无极联合概率数据互联算法

管旭军1,2,周旭1,芮国胜1

1. 海军航空工程学院电子信息工程系，山东烟台 264001;2. 海军湛江保障基地通信雷达声纳修理厂, 广东湛江 524009

出版日期:2009-11-26 发布日期:2010-01-03

Centralized multisensor unscented joint probabilistic data association algorithm

GUAN Xu-jun,ZHOU Xu,RUI Guo-sheng

1. Dept. of Electronic and Information Engineering, Naval Aeronautics and Astronautics Univ., Yantai 264001, China;2. Communication Radar and Sonar Maintenance Depot, Navy Zhanjiang Base, Zhanjiang 524009, China

Online:2009-11-26 Published:2010-01-03

摘要/Abstract

摘要：

针对杂波环境下非线性系统中的多传感器多目标跟踪问题，提出了一种集中式多传感器无极联合概率数据互联算法。该算法中，首先采用无极卡尔曼滤波器实现非线性系统中状态分布的传递，在此基础上应用联合概率数据的思想将单个传感器的量测点迹与航迹互联，最后推广至顺序结构。由于无极卡尔曼滤波器可以获得比扩展卡尔曼滤波算法更高精度的近似，因此能减少非线性模型线性化引起的近似误差对联合概率数据互联概率及状态估计的影响，与基于扩展卡尔曼滤波器思想的顺序多传感器联合概率数据互联算法相比，该算法具有更高的跟踪精度和稳定性，最后通过仿真结果验证了该算法的优越性。

Abstract:

A novel centralized multisensor unscented joint probabilistic data association algorithm, CMSUJPDA, is proposed for the multisensormultitarget tracking problem of nonlinear systems in a clutter environment. In the new algorithm, UKF is used for the propagation of state distribution in the nonlinear system first, then the association of measurement to track is implemented on the principle of JPDA. Based on this, the CMSUJPDA algorithm is derived by use of the sequential MSJPDA technique. Because the approximate accuracy of UKF is higher than EKF, the association probability and state estimation in the proposed algorithm are not affected by the linearization error. Hence the accuracy and robustness of CMSUJPDA are improved compared with the MSJPDA/EKF. Finally the simulation shows the superiority of the new algorithm.

管旭军1,2,周旭1,芮国胜1. 集中式多传感器无极联合概率数据互联算法[J]. Journal of Systems Engineering and Electronics, 2009, 31(11): 2602-2606 .

GUAN Xu-jun,ZHOU Xu,RUI Guo-sheng. Centralized multisensor unscented joint probabilistic data association algorithm[J]. Journal of Systems Engineering and Electronics, 2009, 31(11): 2602-2606 .