基于量测划分的单传感器多目标数据关联算法  

Measurement Assignment Based Single Sensor Multi-Target Data Association Algorithm

作　　者：李林[1,2] 何芳[2] 黄柯棣[1] 华云[3] 刘宝宏[1] 

机构地区：[1]国防科技大学机电工程与自动化学院,长沙410073 [2]中国人民解放军63880部队,洛阳471003 [3]中电集团29所,成都610036

出　　处：《传感技术学报》2008年第11期1872-1875,共4页Chinese Journal of Sensors and Actuators

基　　金：国家自然科学基金项目资助(60704038)

摘　　要：在PDA算法的基础上,提出了一种基于量测划分的单传感器多目标数据关联(MSDA)算法;与JPDA算法相比,MSDA算法采用二维分配算法来消除共有量测对相邻目标的影响;与PDA算法相比,MSDA算法仅采用扩充后的私有量测集合来完成目标航迹的更新过程。对MSDA算法、JPDA算法、PDA算法在不同条件下的关联效果进行了仿真计算。仿真结果表明:在关联精度上,MSDA算法要优于PDA算法,但略低于JPDA算法;在实时性上,MSDA算法与PDA算法基本相当,而要明显优于JPDA算法。Based on PDA algorithm, a new single sensor muitl-target data association （MSDA） algorithm is given. Compared to JPDA algorithm, influence of common measurements is alleviated by two-dimension assignment algorithm in MSDA algorithm. Compared to PDA algorithm, updating of target track only need extended private measurement set in MSDA algorithm. Association performances of MSDA algorithm, JPDA algorithm and PDA algorithm are compared in various tracking environments by using Monte Carlo simulation. Simulation results show computing burden of MSDA algorithm is smaller than JPDA al- gorithm and association accuracy of MSDA algorithm is higher than PDA algorithm.

关 键 词：多目标 数据关联 单传感器 JPDA算法 PDA算法 

分 类 号：TN953[电子电信—信号与信息处理]