用于机动目标跟踪的多模型概率假设密度滤波器  

A Probability Hypothesis Density Filter with Multiple Models for Maneuvering Target Tracking

作　　者：王晓[1] 韩崇昭[1] 

机构地区：[1]西安交通大学电子与信息工程学院,西安710049

出　　处：《西安交通大学学报》2011年第12期1-5,共5页Journal of Xi'an Jiaotong University

基　　金：国家自然科学基金创新研究群体科学基金资助项目(60921003);国家自然科学基金资助项目(61074176);国家"973计划"资助项目(2007CB311006)

摘　　要：针对概率假设密度(PHD)滤波器在多目标跟踪问题中无法解决目标发生较大机动时的目标丢失问题,提出了一种多模型概率假设密度(MM-PHD)滤波器.这种MM-PHD滤波器在粒子PHD滤波器的基础上,使用多模型方法对滤波器中每个描述目标状态的粒子的状态进行更新,再将更新后的粒子代入传统的PHD滤波器中用于估计目标的PHD的分布.该滤波器结合PHD滤波器和多模型方法的特点,可用于目标数未知的多机动目标跟踪,且对目标的数量和状态的估计更加准确.多机动目标跟踪的仿真实验表明,与已有方法相比,该滤波器对目标数的估计与真实情况基本一致,描述多目标状态估计误差的Wasserstein距离值降低了50%以上.A multiple model probability hypothesis density（MM-PHD） filter is proposed to solve the track loss problem of PHD filter in highly maneuvering targets tracking.The MM-PHD filter bases on the particle PHD filter,and updates the state of the particles using the multiple model method.Then the updated state of the particles is used to evaluate the PHD distribution of the targets.The filter combines the features of both the PHD filter and the multiple model method and can deal with multi-targets＇ maneuvering with the number of targets unknown.The proposed filter estimates the number and the state of the targets more accurately than the existing filters do.Simulation results and comparisons with existing algorithms show that the number of targets estimated by the proposed algorithm is consistent with the real situation and the Wasserstein distance value that describes the estimation error decreases more than 50%.

关 键 词：机动目标跟踪 概率假设密度 多模型 估计 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]