基于随机有限集的多目标跟踪技术综述  

Overview of Multi-Target Tracking Technology Based on Random Finite Set

作　　者：严灵杰 顾杰 姜余 徐敏[1,2] 高昭昭 田保立[1,2] 张铁男 YAN Lingjie;GU Jie;JIANG Yu;XU Min;GAO Zhaozhao;TIAN Baoli;ZHANG Tienan(Southwest China Research Institute of Electronic Equipment,Chengdu 610036,China;National Key Laboratory of Electromagnetic Space Security,Chengdu 610036,China;Unit 95438 of the PLA,Meishan 620860,China)

机构地区：[1]中国电子科技集团公司第二十九研究所,成都610036 [2]电磁空间安全全国重点实验室,成都610036 [3]中国人民解放军95438部队,四川眉山620860

出　　处：《电子信息对抗技术》2024年第1期81-88,共8页Electronic Information Warfare Technology

基　　金：国家自然科学基金资助项目(62261059,61963037,61863035)。

摘　　要：随机有限集理论为多目标跟踪、多传感器融合和态势评估等问题提供了完整、统一的理论框架和解决方案。基于随机有限集的跟踪算法将多目标状态和量测建模为随机有限集,自然地引入航迹起始、终结机制,可实现目标数量和状态的同时估计。通过对随机有限集框架下的概率假设密度、带势概率假设密度、多目标多伯努利滤波器、扩展标签随机集滤波器和泊松多伯努利混合滤波器的研究进展进行详细梳理和综合对比,对基于随机有限集的多目标跟踪领域未来发展方向进行了分析和展望。The random finite set theory provides a complete and unified theoretical framework and solution for multi-target tracking,multi-sensor fusion and situation assessment.Multi-target status and measurements are modeled as random finite set by tracking algorithm based on random finite set.By naturally introducing the track initiation and termination mechanism,the simultaneous estimation of target number and status can be achieved.The research progress of probability hypothesis density,cardinalized probability hypothesis density,multi-target multi-Bernoulli filter,generalized label random finite set and Possion multi-Bernoulli hybrid filter under the framework of random finite set is reviewed in detail and compared comprehensively.The future development direction of multi-target tracking based on random finite set is analyzed and prospected.

关 键 词：多目标跟踪 多传感器融合 随机有限集 概率假设密度 带势概率假设密度 

分 类 号：TN953[电子电信—信号与信息处理]