新生目标强度未知的单量测PHD滤波器  

A Single Measurement PHD Filter with Unknown Target Birth Intensity

作　　者：徐从安[1] 熊伟[1] 刘瑜[1] 何友[1] 

机构地区：[1]海军航空工程学院信息融合研究所,山东烟台264001

出　　处：《电子学报》2016年第10期2300-2307,共8页Acta Electronica Sinica

摘　　要：自适应新生目标强度PHD滤波器(PHD-M)在目标漏检时易发生错估或漏估,从而导致滤波器估计性能下降.为解决这一问题,提出了一种新生目标强度未知的单量测(single measurement)PHD滤波器(PHD-SM)并给出了其粒子实现方式.该文首先通过构建一步虚拟量测对漏检目标进行补偿,然后基于单量测PHD分解技术推导了PHD预测和更新公式,最后设计了一种无须聚类操作的多目标状态估计方法.仿真实验表明,在当检测概率PD较小时,PHD-SM滤波器估计性能优于PHD-M滤波器,且检测概率越小,性能优势越明显.In situations where the targets cannot be detected in the surveillance region,the estim ated perform ance of the adaptive target birth intensity probability hypothesis density（ PHD） filter will get worse because of false or lowestim ate. T o overcom e this problem,with unknown target birth intensity,a single m easurem ent PHD（ PHD-S M） filter and its sequential Monte C arlo（ S MC） m ethod are proposed. First,the undetected targets are com pensated through developing the one step virtual m easurem ent set. A fterward,according to the single m easurem ent decom position technique of PHD,the predication and update equations are derived. Finally,a novel m ulti-target state estim ation m ethod is presented. T he sim ulation results showthat,when the detection probability PDis sm all,PHD-S Mfilter has higher estim ation perform ance. Moreover,the sm aller the detection probability,the m ore significant advantage of estim ation perform ance for PHD-S Mfilter.

关 键 词：多目标跟踪 概率假设密度 新生目标强度未知 单量测 一步虚拟量测 

分 类 号：TN957[电子电信—信号与信息处理] TP391[电子电信—信息与通信工程]