基于神经网络的分布式被动传感器信息融合技术

李洪瑞

兵工学报 ›› 2020, Vol. 41 ›› Issue (1) : 95-101.

兵工学报 ›› 2020, Vol. 41 ›› Issue (1) : 95-101. DOI: 10.3969/j.issn.1000-1093.2020.01.011
论文

基于神经网络的分布式被动传感器信息融合技术

  • 李洪瑞
作者信息 +

Neural Network-based Information Fusion Technique for Distributed Passive Sensor

  • LI Hongrui
Author information +
文章历史 +

摘要

在分布式被动传感器信息融合中,存在多传感器信息关联和单传感器目标估计困难,二者相互 依赖和制约,造成相对于不同传感器的信息难于进行时空对准、虚假目标不能消除。为此,应用一种混合式有序分层信息融合结构,避免多传感器信息的多重组合问题,建立了基于两个传感器的信息关联与目标估计联合优化模型,并采取一种优化神经网络算法,避免关联中的组合计算。仿真计算结果表明,这种信息融合结构、优化模型和模拟神经网络的应用是解决被动信息融合系统中关联和估计问题的一种有效方法,所采用的Hopfield型神经网络易于实现,可以提高信息融合的性能。

Abstract

The information correlation (IC) of mult-sensor and the target estimation (TE) of single sensor are difficult in distributed passive sensor (DPS) information fusion (IF). For example, the information from different sensors cannot be registered in time and space, and the false targets cannot be eliminated due to the interdependence and mutual restriction of TE. Therefore a hybrid ordered delaminated information fusion structure (HODIFS) is introduced to avoid the multiple combinations of multi-sensor information. A united optimization model (UOM) based on 2-sensor IC and TE is established, which uses an optimization Hopfield neural network (HNN) algorithm and avoids the complex combination computation of correlation. Simulated results indicate that the HODIFS with UOM based on HNN is effective in the DPSIF, in which HNN is easily realized and the performance of DPSIF can be improved. Key

关键词

分布式被动传感器 / 信息融合 / 关联 / 神经网络

Key words

distributedpassivesensor / informationfusion / correlation / neuralnetwork

引用本文

导出引用
李洪瑞. 基于神经网络的分布式被动传感器信息融合技术. 兵工学报. 2020, 41(1): 95-101 https://doi.org/10.3969/j.issn.1000-1093.2020.01.011
LI Hongrui. Neural Network-based Information Fusion Technique for Distributed Passive Sensor. Acta Armamentarii. 2020, 41(1): 95-101 https://doi.org/10.3969/j.issn.1000-1093.2020.01.011

基金

海军装备预先研究项目(3020102020103)

参考文献



[1]董志荣.舰船信息融合与目标运动分析[M]. 北京:国防工业出版社,2016:114-269.
DONG Z R. Information fusion and target motion analysis for naval vessels[M]. Beijing: National Defense Industry Press, 2016: 114-269. (in Chinese)
[2]李洪瑞, 盛安东.连续纯方位系统的可观测性分析[J]. 兵工学报, 2009, 30(11):1446-1450.
LI H R,SHENG A D. Analysis on the observability for continuous bearings-only system[J]. Acta Armamentarii, 2009, 30(11):1446-1450.(in Chinese)
[3]NARDONES C, AIDALA V J. Observability criteria for bearing-only target motion analysis[J]. IEEE Transactions on Aerospace and Electronic Systems, 1981, 17(2):162-166.
[4]AIDALA V J, HAMMEL S E. Utilization of modified polar coordinates for bearings-only tracking[J]. IEEE Transactions on Automatic Control, 1983, 28(3):283-294.
[5]段广全, 孙书利. 带未知模型参数和衰减观测率系统自校正分布式融合估计[J/OL]. 自动化学报, 2019[2019-03-20]. http:∥kns.cnki.net/kcms/detail/11.2109.TP.20190219.1552.007.html.
DUAN G Q, SUN S L. Self-tuning distributed fusion estimation for systems with unknown model parameters and fading measurement rates[J/OL]. Acta Automatica Sinica, 2019[2019-03-20]. http:∥kns.cnki.net/kcms/detail/11.2109.TP.20190219.1552.007.html.(in Chinese)
[6]XIE Y F, SONG T L. Bearings-only multi-target tracking using an improved labeled multi-Bernoulli filter[J]. Signal Processing, 2018, 151:32-44.
[7]LIN H L, SUN S L. Globally optimal sequential and distributed fusion state estimation for multi-sensor systems with cross-correlated noises[J]. Automatica, 2019, 101:128-137.
[8]李文超, 邹焕新, 雷琳, 等.目标数据关联技术综述[J]. 计算机仿真, 2014, 31(3):1-5,10.
LI W C,ZOU H X,LEIL,et al.A survey of target data association[J]. Computer Simulation, 2014, 31(3):1-5,10.(in Chinese)
[9]SENGUPTAD, ILTIS R A.Neural solution to the multitarget tracking data association problem[J].IEEE Transactions on Aerospace and Electronic Systems, 1989,25(1):96-108.

[10]吕红芳, 顾幸生. 基于蚁群神经网络的两级信息融合算法[J]. 上海交通大学学报, 2016, 50(8):1323-1330.
L H F, GU X S. A two level information fusion algorithm based on ant colony neural network[J]. Journal of Shanghai Jiao Tong University, 2016, 50(8):1323-1330.(in Chinese)
[11]田腾, 尹力, 黄海宁, 等. 基于Hough变换的纯方位目标航迹关联算法[J]. 网络新媒体技术, 2018, 7(2):37-41.
TIAN T,YIN L,HUANG H N, et al. Track correlation algorithm of bearings-only targets based on Hough transform[J]. Microcomputer Applications,2018,7(2):37-41.(in Chinese)
[12]RONALD P S M. 多源多目标统计信息融合[M].范红旗, 卢大成, 刘本源, 等,译. 北京:国防工业出版社, 2013:208-217.
RONALD P S M. Statistical multisource-multitarget information fusion[M]. FAN H Q,LU D C,LIU B Y, et al,translated. Beijing: National Defense Industry Press, 2013: 208-217. (in Chinese)
[13]宋强, 周万宁. 基于最近航迹迭代的目标航迹对准关联算法[J]. 舰船电子工程, 2018, 38(10):69-73.
SONG Q, ZHOU W N. Target track alignment-correlation algorithm based on iterative closet track [J]. Ship Electronic Engineering, 2018, 38(10):69-73.(in Chinese)
[14]衣晓, 张怀巍, 曹昕莹, 等. 基于区间灰数的分布式多目标航迹关联算法[J].航空学报, 2013, 34(2):352-360.
YI X, ZHANG H W,CAO X Y,et al. A track association algorithm for distributed muti-target system based on gray interval numbers[J]. Acta Aeronautica et Astronautica Sinica, 2013, 34(2): 352-360.(in Chinese)
[15]李启虎, 卫翀华, 薛山花. 声呐多传感器观测资料数据融合的一种深度学习算法[J]. 中国科学:信息科学, 2018, 48(12): 1614-1621.
LI Q H, WEI C H, XUE S H. A deep learning algorithm for multiple observation data fusion in sonar system[J].Scientia Sinica:Informationis, 2018, 48(12): 1614-1621. (in Chinese)
[16]李洪瑞, 王超, 肖凯. 纯方位目标运动分析四维状态的最小二乘模型及分析[J]. 指挥控制与仿真, 2007, 29(5):10-14.
LI H R, WANG C, XIAO K. Least square models with 4-dimesional states and analysis for bearings-only TMA[J]. Command Control & Simulation, 2007, 29(5):10-14. (in Chinese)
[17]MATHENDRAM,VIKRAM K,BA-NGU V. 目标跟踪、分类与传感器管理理论及应用[M]. 乔向东, 梁彦, 杨峰, 等, 译. 北京:国防工业出版社, 2017:11-19.
MAHENDRA M,VIKRAM K,BA-NGU V. Integrated tracking, classification and sensor management: theory and application[M]. QIAO X D,LIANG Y,YANG F,et al, translated. Beijing: National Defense Industry Press, 2017: 11-19. (in Chinese)





第41卷第1期
2020年1月兵工学报ACTA
ARMAMENTARIIVol.41No.1Jan.2020

395

Accesses

0

Citation

Detail

段落导航
相关文章

/