Ellipsoidal Tracking of Maneuvering Group Targets Based on GRU-Transformer and Random Matrix

CHEN Ye; LIANG Yuan; LI Yinya; QI Guoqing

doi:10.12382/bgxb.2025.0808

您当前的位置：

首页 >

文章列表页 >

Ellipsoidal Tracking of Maneuvering Group Targets Based on GRU-Transformer and Random Matrix

更新时间：2026-04-24

- Ellipsoidal Tracking of Maneuvering Group Targets Based on GRU-Transformer and Random Matrix
- Acta Armamentarii Vol. 47, Issue 1, Pages: 250808(2026)
- 作者机构：
  
  南京工程学院计算机工程学院，江苏南京 211167
  南京工程学院交通工程学院，江苏南京 211167
  南京理工大学自动化学院，江苏南京 210094
- 作者简介：
- 基金信息：
- DOI：10.12382/bgxb.2025.0808
  CLC： TJ02
- Received：03 September 2025，
  
  Online First：11 February 2026，
  
  Published：31 January 2026
- 稿件说明：
移动端阅览
CHEN Ye, LIANG Yuan, LI Yinya, et al. Ellipsoidal Tracking of Maneuvering Group Targets Based on GRU-Transformer and Random Matrix[J]. Acta Armamentarii, 2026, 47(1): 250808.
DOI：

CHEN Ye, LIANG Yuan, LI Yinya, et al. Ellipsoidal Tracking of Maneuvering Group Targets Based on GRU-Transformer and Random Matrix[J]. Acta Armamentarii, 2026, 47(1): 250808. DOI： 10.12382/bgxb.2025.0808.

摘要

针对机动群目标跟踪问题，提出基于多任务多头门控循环单元变换器（Gated Recurrent Unit Transformer

GRU-Transformer）的机动模型辨识与当前统计（Current Statistical

CS）模型机动频率参数

回归，并将判别结果融入随机矩阵椭球-贝叶斯更新，实现群目标质心与扩展外形的联合跟踪。基于机动群目标时间序列量测数据，提取群目标运动多项特征参数，输入所提深度神经网络，实现机动群目标运动模型（匀速（Constant Velocity

CV）模型、匀加速（Constant Acceleration

CA）模型、CS模型）的精准辨别，若目标模型为CS运动模型，同时输出机动频率参数

的精确估计结果。运用随机矩阵群目标跟踪理论，提出一种基于贝叶斯状态估计架构的椭圆机动群目标跟踪方法。仿真实验结果分析表明：新方法可实现对机动椭圆群目标的精确稳健跟踪，相较于传统交互式多模型群目标跟踪方法，跟踪精度有显著提升。

Abstract

For the tracking of maneuvering group target

this paper proposes a multi-task multi-head gated recurrent unit (GRU)-Transformer for the dynamic model identification and the regression of current statistical (CS) model maneuver frequency

. The inferred model probabilities and estimated

are then incorporated into a random-matrix ellipsoidal Bayesian update to jointly track the centroid and extended shape of group target. The multiple motion feature parameters of group target are extracted from time-series measurement data

then inputted into a deep neural network (DNN )

achieving the accurate discrimination of maneuvering group target constant velocity (CV) model

constant acceleration (CA) model

and CS model. For the CS model

the DNN further outputs the precise estimates of the maneuvering frequency parameter

. On this basis

the random matrix theory is employed to develop an elliptical maneuvering group target tracking method based on Bayesian state estimation framework. Simulated results demonstrate that the proposed method achieves the accurate and robust tracking of maneuvering elliptical group targets

and significantly improves the tracking accuracy compared with traditional interacting multiple model approaches.

关键词

Keywords

references

才华，周鸿策，付强，等.基于多层特征嵌入的单目标跟踪算法[J].兵工学报，2025,46（3）:240062.

CAI H, ZHOU H C, FU Q, et al.Single object tracking algorithm based on multilayer feature embedding[J].Acta Armamentarii, 2025,46（3）:240062.（in Chinese）

邹汝平，刘建书.基于概率假设密度滤波与无迹Kalman滤波的多目标跟踪与识别[J].兵工学报，2020,41（8）:1502-1508.

ZOU R P, LIU J S.Multi-target tracking and recognition technology based on PHD and UKF[J].Acta Armamentarii,2020,41（8）:1502-1508.（in Chinese）

LAN J, LI X R. Extended-object or group-target tracking using random matrix with nonlinear measurements[J]. IEEE Transactions on Signal Processing,2019,67（19）:5130-5142.

LIANG Z N, YAN Z H, GAO M, et al. Three-dimensional group target separation detection method based on ellipsoid shape reconstruction[J]. IEEE Transactions on Radar Systems,2024,2:767-777.

TUNCER B, ORGUNER U,ÖZKAN E. Multi-ellipsoidal extended target tracking with variational Bayes inference [J]. IEEE Transactions on Signal Processing,2022,70:3921-3934.

KOCH J W. Bayesian approach to extended object and cluster tracking using random matrices [J]. IEEE Transactions on Aerospace and Electronic Systems,2008,44（3）:1042-1059.

LAN J, LI X R. Tracking of maneuvering non-ellipsoidal extended object or target group using random matrix[J]. IEEE Transactions on Signal Processing,2014,62（9）:2450-2463.

HOHER P, BAUR T, REUTER J, et al. 3D-extended object tracking and shape classification with a lidar sensor using random matrices and virtual measurement models[C]∥Proceedings of the International Conference on Information Fusion. Venice, Italy:IEEE,2024:1-8.

ALQADERI H, GOVAERS F, SCHULZ R. Spacial elliptical model for extended target tracking using laser measurements [C]∥Proceedings of the Sensor Data Fusion: Trends, Solutions, Applications. Bonn, Germany:IEEE,2019:1-6.

JIANG Q, WANG R, ZHANG J C, et al. A multi-subobject approach to dynamic formation target tracking using random matrices[J]. IEEE Transactions on Aerospace and Electronic Systems,2023,59（6）:7334-7351.

ZHANG L, LAN J. Extended object tracking using random matrix with skewness [J]. IEEE Transactions on Signal Processing, 2020,68:5107-5121.

JIAO Q Q, YANG X J. Distributed variational measurement update for extended target tracking with random matrix[J]. IEEE Transactions on Aerospace and Electronic Systems,2024,60（4）:3792-3806.

陈辉，边斌超，连峰，等.基于Transformer复杂运动辨识的机动星凸形扩展目标跟踪方法[J].雷达学报（中英文）,2024, 13（3）:629-645.

CHEN H, BIAN B C, LIAN F, et al.A novel method for tracking complex maneuvering star convex extended targets using transformer network[J].Journal of Radars,2024,13（3）:629-645.（in Chinese）

CHEN Y D, JIAO Y W, WU M, et al. Group target tracking for highly maneuverable unmanned aerial vehicles swarms: a perspective[J]. Sensors,2023,23（9）:4465.

THARANI R, AMIRALI K G, REZA H. Geometrically-informed sequential monte carlo method for dynamic swarm tracking[J]. ISA Transactions,2025,164:149-160.

吴孙勇，周于松，谢芸，等.基于MM-GGIW-PMBM滤波的扩展目标跟踪算法[J].北京航空航天大学学报，2022,48（12）:2356-2364.

WU S Y, ZHOU Y S, XIE Y, et al.Extended target tracking algorithm based on MM-GGIW-PMBM filter [J].Journal of Beijing University of Aeronautics and Astronautics, 2022, 48（12）:2356-2364.（in Chinese）

WANG S H, MEN C K, LI R X, et al. A maneuvering extended target tracking IMM algorithm based on second-order EKF[J]. IEEE Transactions on Instrumentation and Measurement, 2024, 73:1-11.

HUANG J J, HU H Q, KANG L. Time convolutional network-based maneuvering target tracking with azimuth-Doppler measurement[J]. Sensors,2024,24（1）:263.

DI Y, LI R H, TIAN H, et al. A maneuvering target tracking based on fast IMM-extended Viterbi algorithm[J]. Neural Computing &Applications,2025,37:7925-7934.

程稳，陈忠碧，李庆庆，等.时空特征对齐的多目标跟踪算法[J].光电工程，2023,50（6）:68-81.

CHENG W, CHEN Z B, LI Q Q, et al.Multiple object tracking with aligned spatial-temporal feature [J].Opto-Electronic Engineering,2023,50（6）:68-81.（in Chinese）

刘超军，段喜萍，谢宝文.应用GhostNet卷积特征的ECO目标跟踪算法改进[J].激光技术，2022,46（2）:239-247.

LIU C J, DUAN X P, XIE B W.Improvement of ECO target tracking algorithm based on GhostNet convolution feature[J].Laser Technology,2022,46（2）:239-247.（in Chinese）

吴非，张建林.结合全局光流的孪生区域提名网络目标跟踪算法[J].半导体光电，2023,44（3）:422-428.

WU F, ZHANG J L.Siamese region proposal network object tracking algorithm with global optical flow [J].Semiconductor Optoelectronics,2023,44（3）:422-428.（in Chinese）

LIU J X, WANG Z L, XU M. DeepMTT: a deep learning maneuvering target-tracking algorithm based on bidirectional LSTM network[J].Information Fusion,2020,53:289-304.

LI Z X, WANG Y, LIU J M, et al. Variable-structure interacting multiple-model estimation for group targets tracking with random matrices[J]. Journal of Aerospace Engineering,2018,232（7）:1201-1211.

Views

下载量

CNKI被引量

Alert me when the article has been cited

提交

Tools

Publicity Resources

No data

Related Author

CHEN Ye

LIANG Yuan

LI Yinya

QI Guoqing

Related Institution

School of Computer Engineering, Nanjing Institute of Technology

School of Traffic Engineering, Nanjing Institute of Technology

School of Automation, Nanjing University of Science and Technology

Postal code：100089
Tel：010-68963060/68962718 Fax：010-68963025 Email：bgxb@cos.org.cn
Technical support is provided by Beijing Founder electronics co., LTD 京ICP备05059581号-4 京公网安备11010802024360号
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
Total Visits：119049 Visits Today：36

⁰