杂波环境下基于最大熵模糊聚类的JPDA算法  

JPDA algorithm based on maximum entropy fuzzy clustering in clutter environment

作　　者：毕文豪 周杰 张安[1] 刘力 BI Wenhao;ZHOU Jie;ZHANG An;LIU Li(School of Aeronautics,Northwestern Polytechnical University,Xi’an 710072,China;Nanjing Institute of Electronic Technology,Nanjing 210039,China)

机构地区：[1]西北工业大学航空学院,陕西西安710072 [2]南京电子技术研究所,江苏南京210039

出　　处：《系统工程与电子技术》2023年第7期1920-1927,共8页Systems Engineering and Electronics

基　　金：国家自然科学基金(61903305,62073267);航空科学基金(201905053001)资助课题。

摘　　要：针对杂波环境下的多目标跟踪数据关联存在跟踪精度低、实时性差的问题,提出了一种基于最大熵模糊聚类的联合概率数据关联算法(joint probabilistic data association algorithm based on maximum entropy fuzzy clustering,MEFC-JPDA)。首先,采用最大熵模糊聚类求得的隶属度初步表征目标与有效量测之间的关联概率。其次,采用基于目标距离的量测修正因子对关联概率进行调整,并建立关联概率矩阵。最后,结合卡尔曼滤波算法,对目标的状态进行加权更新。仿真结果表明,所提算法在杂波环境下的跟踪性能相比现有的两种关联算法有较大提升,是一种有效的多目标跟踪数据关联算法。Aiming at the problems of low tracking accuracy and poor real-time performance of multi-target tracking data association in clutter environment,this paper proposes a joint probabilistic data association algorithm based on maximum entropy fuzzy clustering(MEFC-JPDA).Firstly,the membership obtained by the maximum entropy fuzzy clustering is used to preliminarily characterize the correlation probability between the target and the effective measurement.Secondly,the measurement correction factor based on target distance is used to adjust the correlation probability,and the correlation probability matrix is established.Finally,combined with the Kalman filtering algorithm,the state of the target is weighted updated.Simulation results show that the tracking performance of the proposed algorithm in clutter environment is greatly improved compared with the existing two association algorithms,and it is an effective multi-target tracking data association algorithm.

关 键 词：多目标跟踪 联合概率数据关联 最大熵模糊聚类 量测修正因子 

分 类 号：TN953[电子电信—信号与信息处理]