Multitarget-tracking Method for Airborne Radar Based on a Transformer Network

被引：0
|
作者：
Li W. [1 ]
Zhang S. [1 ]
Wang W. [2 ]
机构：
[1] Research Institute of Electronic Science and Technology, University of Science and Technology of China, Chengdu
[2] School of Information and Communication Engineering, University of Science and Technology of China, Chengdu
来源：
Journal of Radars | 2022年 / 11卷 / 03期
基金：
中国国家自然科学基金;
关键词：
Airborne radar; Attention mechanism; Data association; Multitarget-tracking; Transformer network;
D O I：
10.12000/JR22009
中图分类号：
TN95 [雷达];
学科分类号：
080904 ; 0810 ; 081001 ; 081002 ; 081105 ; 0825 ;
摘要：
Conventional multitarget-tracking data association algorithms must have prior information, such as the target motion model and clutter density. However, such prior information cannot be obtained timely and accurately before tracking. To address this issue, a data association algorithm for multitarget tracking based on a transformer network is proposed. First, considering that the radar may not perform accurate detected the target, virtual measurements are performed to re-establish the data association model. Thus, a data association method based on the transformer network is proposed to solve the matching problem of multitargets and multimeasurements. Moreover, a loss function combining Masked Cross entropy loss and Dice (MCD) loss is designed to optimize the network parameters. Simulation data and real measurement data results show that the proposed algorithm outperforms classic data association algorithms and algorithms based on bidirectional long short-term memory network under varying detection probability conditions. © 2022 Institute of Electronics Chinese Academy of Sciences. All rights reserved.
引用
收藏
页码：469 / 478
页数：9
相关论文
共 24 条
  • [1] BERTSIMAS D, SAUNDERS Z, SHTERN S., Multitarget tracking via mixed integer optimization[J], IEEE Transactions on Automatic Control, 63, 11, pp. 3627-3642, (2018)
  • [2] EMAMI P, PARDALOS P M, ELEFTERIADOU L, Et al., Machine learning methods for solving assignment problem in multi-object tracking, (2018)
  • [3] WANG Jianguo, HE Peikun, CAO Wei, Study on the Hungarian algorithm for the maximum likelihood data association problem[J], Journal of Systems Engineering and Electronics, 18, 1, pp. 27-32, (2007)
  • [4] ZHENG Le, WANG Xiaodong, Improved multiple hypothesis tracker for joint multiple target tracking and feature extraction[J], IEEE Transactions on Aerospace and Electronic Systems, 55, 6, pp. 3080-3089, (2019)
  • [5] ZHANG Guangnan, LIU Penghui, Probabilistic data association algorithm based on modified input estimation[C], 2011 7th International Conference on Wireless Communications, Networking and Mobile Computing, pp. 1-4, (2011)
  • [6] Longqiang NI, Shesheng GAO, Li XUE, Improved probabilistic data association and its application for target tracking in clutter[C], 2011 International Conference on Electronics, Communications and Control (ICECC), pp. 293-296, (2011)
  • [7] WANG Yuhuan, WANG Jinkuan, WANG Bin, A modified multi-target tracking algorithm based on joint probability data association and Gaussian particle filter[C], The 11th World Congress on Intelligent Control and Automation, pp. 2500-2504, (2014)
  • [8] Shaoming HE, SHIN H S, TSOURDOS A., Joint probabilistic data association filter with unknown detection probability and clutter rate[C], 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI), pp. 559-564, (2017)
  • [9] AINSLEIGH P L, LUGINBUHL T E, WILLETT P K., A sequential target existence statistic for joint probabilistic data association[J], IEEE Transactions on Aerospace and Electronic Systems, 57, 1, pp. 371-381, (2021)
  • [10] HOPFIELD J J, TANK D W., “Neural” computation of decisions in optimization problems[J], Biological Cybernetics, 52, 3, pp. 141-152, (1985)
123