一种基于粒子滤波的被动多传感器多目标跟踪算法  

Multiply Targets Tracking Algorithm Based on Particle Filter in Multiple Passive Sensors System

作　　者：赵欣[1,2] 姬红兵[1] 杨柏胜[1] 

机构地区：[1]西安电子科技大学电子工程学院,西安710071 [2]中国航天科技集团第四研究院401所,西安710025

出　　处：《系统仿真学报》2009年第20期6382-6386,6390,共6页Journal of System Simulation

基　　金：国家自然科学基金(60677040)

摘　　要：针对被动观测系统中非线性运动多目标跟踪问题,提出了一种基于交叉定位的模糊-概率双加权粒子滤波跟踪算法。算法利用多个被动传感器的角度观测信息进行交叉定位,得到目标的位置观测信息,通过模糊-概率双加权完成目标与定位点的关联匹配,最后利用粒子滤波对非线性运动的目标进行跟踪;其中关联算法和滤波算法的有效结合是该算法的创新点。仿真结果表明,所提出的算法可以准确地排除虚假定位点,可有效跟踪多个非线性运动目标。For non-linear and multi-target tracking problem based on multiple passive sensors system, a novel particle filter based on fyzzy-probablity weighting algorithm was proposed. In this algorithm, it was proposed to estimate the intersection location points by the least square algorithm and use the fuzzy-probability weighting data association algorithm to get available association for multiple targets. Then, a particle filter was adopted to track targets with nonlinear motion. The emphasis was placed on the effective combination of the association model with filter model. The simulation results show that the proposed algorithm can achieve higher tracking precision and effective elimination of the ghost points in tracking multiple targets moving nonlinearly by passive sensors.

关 键 词：被动观测 多目标跟踪 模糊关联度 粒子滤波 

分 类 号：TN953[电子电信—信号与信息处理]