基于变分贝叶斯联合优化的情报监视与侦察  

Intelligence,Surveillance and Reconnaissance Based on Variational Bayesian Joint Optimization

作　　者：潘泉[1,2] 胡玉梅[1,2] 马季容 PAN Quan;HU Yumei;MA Jirong(School of Automation,Northwestern Polytechnical University,Xi'an 710072,China;Key Laboratory of Information Fusion Technology,Ministry of Education,Xi'an 710072,China;Xi'an Advanced Control Technologies Research Institute,Xi'an 710065,China)

机构地区：[1]西北工业大学自动化学院,西安710072 [2]西北工业大学信息融合教育部重点实验室,西安710072 [3]西安现代控制技术研究所,西安710065

出　　处：《指挥信息系统与技术》2020年第2期1-8,共8页Command Information System and Technology

基　　金：国家自然科学基金(61790552和61374159);中国科协优秀中外青年交流计划(2017CASTQNJL046);西北工业大学博士论文创新(CX201915)资助项目。

摘　　要：通过分析情报监视与侦察系统中亟待解决的目标机动、群体目标、目标存在性、量测多路径、传感器组网融合、高维数、深耦合以及不可避免的系统非线性等问题,在变分贝叶斯统一优化框架下给出了解决上述问题的联合优化方法的一般形式。以天波超视距雷达组网融合的远程预警系统为例,讨论了多机动目标的航迹管理、数据关联、参数辨识和状态估计问题。最后,探讨了情报监视与侦察领域的开放问题及研究方向。The problems of target maneuver,multi-targets,existence of target,multi-paths,netted sensors fusion,high-dimensionality,deep coupling,and the inevitable nonlinearity are urgent to be solved for the intelligence,surveillance and reconnaissance systems.By analyzing above problems,a general joint optimization method is given based on the framework of variational Bayesian unified optimization.Furthermore,taking the long-range early warning system of netted over-the-horizon radars as an example,the track management,data association,parameter identification,and state estimation of multiple maneuvering targets are discussed.Finally,the open issues and future directions in the field of intelligence,surveillance and reconnaissance are discussed.

关 键 词：情报监视与侦察 联合优化 变分贝叶斯 目标跟踪 天波超视距雷达 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]