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.
出处
《现代电子技术》
2012年第4期18-21,24,共5页
Modern Electronics Technique
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
multi-target tracking
feature aided tracking
generalized probability data association
dense clutter
