多目标跟踪中基于特征辅助的概率数据关联算法  

Algorithm base on feature assist for probability data association in multi-target tracking

作　　者：马璐[1] 王刚 

机构地区：[1]北京市遥感信息研究所,北京100192 [2]北京环球信息开发应用中心,北京100094

出　　处：《现代电子技术》2012年第4期18-21,24,共5页Modern Electronics Technique

摘　　要：在传统的多目标跟踪系统中,数据关联仅利用了那些与目标状态向量直接相关的信息。在此提出了一种基于广义概率数据关联(GPDA)的新的关联算法即特征辅助跟踪(FAT)算法。该算法同时利用了目标的特征信息和状态信息进行数据关联,较好地解决了在密集杂波环境下对近目标的跟踪问题。最后以目标的一维距离像信息为例进行仿真,仿真结果表明,所提出的算法使跟踪性能优于传统的概率数据关联。In traditional multi-target tracking systems,the information only relative to target state vector has been used for data association.A new association algorithm based on the generalized probability data association(GPDA) algorithm-feature aided tracking(FAT) algorithm is presented in this paper.FAT algorithm combines feature information with traditional state information in a probabilistic way.It preferably solves the tracking problem of closely spaced targets in dense clutter.The 1D range profile information of targets is taken as an example to perform a simulation.The simulation results verifies that the FAT algorithm outperforms the conventional probability data association algorithm.

关 键 词：多目标跟踪 特征辅助跟踪 广义概率数据关联 密集杂波 

分 类 号：TN95[电子电信—信号与信息处理]