基于部分可观马氏决策过程的多平台主被动传感器调度

doi:10.3969/j.issn.0372-2112.2014.10.037

电子学报 ›› 2014, Vol. 42 ›› Issue (10): 2104-2109.DOI: 10.3969/j.issn.0372-2112.2014.10.037

基于部分可观马氏决策过程的多平台主被动传感器调度

张子宁^1,2, 单甘霖¹, 段修生¹

1. 军械工程学院电子与光学工程系, 河北石家庄 050003;
2. 北京航天飞控中心, 北京 100094

收稿日期:2013-06-17 修回日期:2014-04-09 出版日期:2014-10-25
- 作者简介:
- 张子宁 男,1984年生于河北石家庄,博士,主要从事传感器管理及信息融合方面的研究. E-mail:iron_zay@126.com;单甘霖 男,1962年生于江苏如东,教授,主要从事信息融合、神经网络和目标跟踪等方面的研究工作.段修生男,1970年生于河北,副教授,主要研究方向为人工智能和故障诊断.
- 基金资助:
- 军内科研重点项目资助课题

POMDP-Based Scheduling of Active/Passive Sensors in Multi-Platform

ZHANG Zi-ning^1,2, SHAN Gan-lin¹, DUAN Xiu-sheng¹

1. Department of Electronic and Optical Engineering, Ordnance Engineering College, Shijiazhuang, Hebei 050003, China;
2. Beijing Aerospace Control Center, Beijing 100094, China

Received:2013-06-17 Revised:2014-04-09 Online:2014-10-25 Published:2014-10-25
- Supported by:
- Key scientific research projects within the military

摘要/Abstract

摘要：

为了使有限时域内的跟踪精度和辐射风险达到最佳平衡,本文研究了多传感器平台在协同跟踪目标时的主被动传感器调度问题.将该问题建立成基于部分可观马氏决策过程的数学模型以同步实现目标跟踪和辐射控制.在先见优化思想的基础上,借助由无迹采样近似得到的精度收益及由隐马氏模型滤波器推导出的辐射代价将调度问题转化成决策树问题,并采用分枝定界方法求解.仿真结果表明了该方法的有效性.

关键词: 传感器调度, 部分可观马氏决策过程, 先见优化, 无迹采样, 分枝定界

Abstract:

To make an optimal trade-off between the tracking accuracy and the radiation risk in a period of time, this paper studies the scheduling problem of selecting the active/passive sensors in the multi-platform for target tracking.The problem is formulated as a partially observable Markov decision process (POMDP), which can take both target tracking and emission control into account.Based on the foresight optimization, the approximate accuracy reward and the radiation cost, which are derived from the unscented transformation sampling and hidden Markov model (HMM) filter respectively, transform our problem into a tree search problem, and the branch and bound method is used for problem solution.The simulation results demonstrate the effectiveness of our approach.

Key words: sensor scheduling, partially observable Markov decision process, foresight optimization, unscented transformation sampling, branch and bound

中图分类号:

TP212.9

张子宁, 单甘霖, 段修生. 基于部分可观马氏决策过程的多平台主被动传感器调度[J]. 电子学报, 2014, 42(10): 2104-2109.

ZHANG Zi-ning, SHAN Gan-lin, DUAN Xiu-sheng . POMDP-Based Scheduling of Active/Passive Sensors in Multi-Platform[J]. Acta Electronica Sinica, 2014, 42(10): 2104-2109.

参考文献

[1] C Kreucher,K Kastella,A O Hero.A Bayesian method for integrated multitarget tracking and sensor management[A].Proceedings of International Conference on Information Fusion[C].Cairns,Qld,Australia:IEEE,2003.132-139.
[2] 周文辉,胡卫东,等.基于协方差控制的集中式传感器分配算法研究[J].电子学报,2003,31(12):2158-2162. Zhou Wen-hui,Hu Wei-dong,et al.A study of sensor allocation algorithm based on covariance control in centralized multisensor systems[J].Acta Electronica Sinica,2003,31(12):2158-2162.(in Chinese)
[3] A Logothetis,A Isaksson.On sensor scheduling via information theoretic criteria[A].Proceedings of the American Control Conference[C].San Diego,CA,USA:American Automatic Control Council,1999.2402-2406.
[4] 程洪玮,等.一种基于信息决策树的低轨星座传感器调度方法[J].电子学报,2010,38(11):2630-2634. Cheng Hong-wei,et al.A sensor scheduling method of LEO constellation based on information decision tree[J].Acta Electronica Sinica,2010,38(11):2630-2634.(in Chinese)
[5] Y Li,L W Krakow,E P K Chong,et al.Approximate stochastic dynamic programming for sensor scheduling to track multiple target[J].Digital Signal Processing,2009,19(6):987-989.
[6] Y He,E P K Chong.Sensor scheduling for target tracking:A Monte Carlo sampling approach[J].Digital Signal Processing,2006,16(5):533-545.
[7] L Kaelbling,M Littman,A Cassandra.Planning and acting in partially observable stochastic domains[J].Artificial Intelligence,1998,101(1-2):99-134.
[8] V Krishnamurthy.Emission management for low probability intercept sensors in network centric warfare[J].IEEE Transactions on Aerospace and Electronic Systems,2005,40(1):133-152.
[9] T L Song,J Y Ahn,C Park.Suboptimal filter design with pseudomeasurements for targets tracking[J].IEEE Transactions on Aerospace and Electronic Systems,1988,24(1):28-39.
[10] X R Li,V P Jilkov.A survey ofmaneuvering target tracking-part III:measurement models[A].Proceedings of SPIE Conference on Signal and Data Processing of Small Targets[C].San Diego,CA,USA:SPIE,2001.559-583.
[11] 刘敏华,萧德云.基于相似度的多传感器数据融合[J].控制与决策,2004,19(5):534-537. Liu Min-hua,Xiao De-yun.Multi-sensor data fusion based on similitude degree[J].Decision and Control,2004,19(5):534-537.(in Chinese)
[12] S J Julier,J K Uhlmann.Unscented filtering and nonlinear estimation[J].Proceedings of the IEEE,2004,92(3):401-422.
[13] L R Rabiner.A tutorial on hidden Markov models and selected applications in speech recognition[J].Proceedings of the IEEE,1989,77(2):257-285.
[14] Z G Feng,K L Teo,V Rehbock.Hybrid method for a general optimal sensor scheduling problem in discrete time[J].Automatic,2008,44(5):1295-1303.

基于部分可观马氏决策过程的多平台主被动传感器调度

POMDP-Based Scheduling of Active/Passive Sensors in Multi-Platform

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

[1]	罗开平;姜维;李一军. 传感器管理述评[J]. 电子学报, 2010, 38(8): 1900-1907.
[2]	赵莹;张健沛;杨静;王冠军. 一种改进的分枝定界半监督支持向量机学习算法[J]. 电子学报, 2010, 38(2): 449-454.