基于多尺度变换的多模态机动多目标航迹检测方法

宋燕; 李天成; 王静远; 李固冲; 苗昊春

doi:10.12263/DZXB.20260715

您当前的位置：

首页 >

文章列表页 >

基于多尺度变换的多模态机动多目标航迹检测方法

更新时间：2026-07-31

- 基于多尺度变换的多模态机动多目标航迹检测方法
- Multiscale Transform for Multimodal Maneuvering Multi-Target Trajectory Detection
- 电子学报 2026年页码：1-19
- 作者机构：
  
  1.西北工业大学自动化学院，陕西西安 710129
  2.西安现代控制技术研究所，陕西西安 710065
- 作者简介：
  
  宋燕女，1999年8月出生于山西省长治市。现为西北工业大学自动化学院博士生。主要研究方向为雷达目标跟踪、多源信息融合。E-mail: syzx@mail.nwpu.edu.cn
  李天成男，1986年6月出生于山东省菏泽市。现为西北工业大学自动化学院教授。主要研究方向为目标跟踪、多传感器信息融合和智能估计等。E-mail: t.c.li@nwpu.edu.cn
  王静远女，1998年5月出生于陕西省汉中市。现为西北工业大学自动化学院博士生。主要研究方向为雷达目标跟踪、多源信息融合。E-mail: jy_wang@mail.nwpu.edu.cn
  李固冲男，1992年2月出生于河南省商丘市。现为西北工业大学副教授。主要研究方向为随机有限集、目标跟踪和多传感器数据融合。E-mail: guchong.li@nwpu.edu.cn
  苗昊春男，1986年6月出生于内蒙古自治区赤峰市。现为西安现代控制技术研究所教授级高级工程师。主要研究方向为制导控制、飞行器控制、协同制导与复杂系统控制等。E-mail: 357285772@qq.com
- 基金信息：
  
  国家自然科学基金(62422117;U2541215;62471400)
- DOI：10.12263/DZXB.20260715
  中图分类号： TP957.51;
- 收稿：2026-06-12，
  
  录用：2026-07-02，
  
  网络首发：2026-07-31，
- 稿件说明：
移动端阅览
宋燕, 李天成, 王静远, 等. 基于多尺度变换的多模态机动多目标航迹检测方法[J/OL]. 电子学报, 2026,1-19.

SONG Yan, LI Tiancheng, WANG Jingyuan, et al. Multiscale Transform for Multimodal Maneuvering Multi-Target Trajectory Detection[J/OL]. ACTA ELECTRONICA SINICA, 2026, 1-19.
宋燕, 李天成, 王静远, 等. 基于多尺度变换的多模态机动多目标航迹检测方法[J/OL]. 电子学报, 2026,1-19. DOI： 10.12263/DZXB.20260715.

SONG Yan, LI Tiancheng, WANG Jingyuan, et al. Multiscale Transform for Multimodal Maneuvering Multi-Target Trajectory Detection[J/OL]. ACTA ELECTRONICA SINICA, 2026, 1-19. DOI： 10.12263/DZXB.20260715.

摘要

针对目标数目未知、高速机动且存在杂波干扰和量测噪声的复杂密集目标探测与跟踪问题，本文提出了一种基于多尺度变换的多模态机动多目标航迹检测方法。现有基于显式数据关联和预设运动模型的多目标跟踪方法，在目标数目时变、机动模式差异显著以及密集杂波干扰条件下，通常需要同时处理目标数量判断、量测关联、模型选择与状态估计等多个相互耦合的问题，数学上是一个NP-hard难题，面临关联假设空间膨胀、模型失配和计算开销增大等挑战。为此，本文基于连续时间航迹函数（Trajectory Function of Time，T-FoT）框架，将滑动时间窗内目标运动表示为由有限维参数刻画的连续时间函数，并构建多尺度航迹参数空间，其中不同尺度对应匀速、加速和局部转弯等不同复杂度的T-FoT航迹族。通过将时间滑窗内量测数据投影至各种候选尺度参数空间，目标航迹起始与检测问题被转化为参数空间内量测投影一致性的检测问题：同源量测在相应尺度下会形成局部聚集的支持响应，而杂波量测或尺度不匹配量测通常难以形成稳定聚集。进一步，针对硬支持计数难以区分一致性量测贡献强弱的问题，提出加权支持累积机制，根据候选航迹与量测之间的一致性残差分配支持权重，增强参数空间中的一致性表征能力，从而提高真实航迹峰与伪峰之间的可分性。在算法实现上，所提方法结合随机最小子集采样、局部峰值筛选、非极大值抑制和跨尺度候选融合，在不显式枚举量测关联关系和目标数量假设的情况下提取高支持航迹候选，并为后续递推跟踪提供初始化信息。仿真结果表明，所提方法在白噪声和有色量测噪声两种条件下均能够保持较高的航迹起始成功率和较低的虚假航迹率；在航迹估计阶段，该方法有效平衡估计精度和计算复杂度，可支撑事后航迹复原、运动过程重建及态势分析等复杂任务。

Abstract

To address the challenging detection and tracking of densely distributed targets in complex scenarios characterized by an unknown number of targets

high-speed maneuvers

clutter

and measurement noise

this paper proposes a multimodal maneuvering multi-target trajectory detection method based on a multiscale transform. Existing multi-target tracking methods based on explicit data association and predefined motion models generally need to deal simultaneously with several mutually coupled tasks

including determining the number of targets

associating measurements with targets

selecting appropriate motion models

and estimating target states. When the number of targets varies over time

the maneuvering modes of different targets differ significantly

and the environment contains dense clutter

these coupled tasks become particularly difficult to handle. Mathematically

the resulting coupled optimization problem is NP-hard

leading to challenges such as the rapid expansion of the association hypothesis space

motion-model mismatch

and increased computational cost. To address these challenges

the proposed method is developed within the framework of the trajectory function of time(T-FoT). Within a sliding time-window

the motion of each target is represented as a continuous-time function characterized by a finite-dimensional parameter vector

on the basis of which multiscale trajectory parameter spaces are constructed. Different scales correspond to T-FoT trajectory families with different levels of complexity

including constant-velocity motion

accelerated motion

and local turning motion

thereby enabling motion patterns of different complexity to be represented within the same multiscale framework. Measurements collected within the sliding time-window are projected into the parameter spaces associated with the various candidate scales. In this manner

the original problems of target trajectory initiation and detection are reformulated as the task of detecting measurement-projection consistency in the corresponding parameter spaces. Measurements originating from the same target tend to produce locally aggregated support responses at the corresponding scale

whereas clutter measurements or measurements projected at a mismatched scale generally fail to form stable support aggregation. The local aggregation of support responses therefore provides the basis for identifying potential target trajectories in the parameter spaces. Conventional hard support counting

however

treats all measurements that satisfy a consistency condition as making equal contributions and therefore cannot distinguish the different contribution strengths of consistent measurements. To overcome this limitation

a weighted support accumulation mechanism is proposed. The mechanism assigns support weights according to the consistency residuals between candidate trajectories and measurements

thereby enhancing the representation of measurement consistency in the parameter space and improving the separability between true trajectory peaks and false peaks. For algorithmic implementation

the proposed method combines random minimal-subset sampling

local peak selection

non-maximum suppression

and cross-scale candidate fusion. Consequently

high-support trajectory candidates can be extracted without explicitly enumerating measurement associations or hypotheses regarding the number of targets

and the extracted candidates can provide initialization information for subsequent recursive tracking. Simulation results show that the proposed method maintains a high trajectory initiation success rate and a low false trajectory rate under both white and colored measurement noise. In the trajectory estimation stage

the method effectively balances estimation accuracy and computational complexity. The proposed method can therefore support complex tasks such as retrospective trajectory recovery

motion process reconstruction

and situation analysis.

关键词

Keywords

references

Wang Jinhao , Chen Xiaolong , Guan Jian , et al . A time-frequency representation method based on ETF-MDNet for radar target micro-motion features [J ] . Chinese Journal of Electronics , 2025 , 34 ( 4 ): 1199 - 1208 . DOI: 10.23919/cje.2024.00.233 http://dx.doi.org/10.23919/cje.2024.00.233

Bar-Shalom Y , Li X R , Kirubarajan T . Estimation with Applications to Tracking and Navigation: Theory, Algorithms and Software [M ] . New York, USA : John Wiley & Sons , 2002 . DOI: 10.1002/0471221279 http://dx.doi.org/10.1002/0471221279

李天成 , 谢昱昕 , 李固冲 , 等 . 面向多目标跟踪的数据关联方法研究综述 [J ] . 雷达科学与技术 , 2025 , 23 ( 1 ): 10 - 31 . DOI: 10.3969/j.issn.1672-2337.2025.01.002 http://dx.doi.org/10.3969/j.issn.1672-2337.2025.01.002

Li Tiancheng , Xie Yuxin , Li Guchong , et al . A survey on data association approaches to multi-target tracking [J ] . Radar Science and Technology , 2025 , 23 ( 1 ): 10 - 31 . (in Chinese) . DOI: 10.3969/j.issn.1672-2337.2025.01.002 http://dx.doi.org/10.3969/j.issn.1672-2337.2025.01.002

Yan Bo , Paolini E , Xu Luping , et al . A target detection and tracking method for multiple radar systems [J ] . IEEE Transactions on Geoscience and Remote Sensing , 2022 , 60 : 1 - 21 . DOI: 10.1109/tgrs.2022.3183387 http://dx.doi.org/10.1109/tgrs.2022.3183387

Fortmann T , Bar-Shalom Y , Scheffe M . Sonar tracking of multiple targets using joint probabilistic data association [J ] . IEEE Journal of Oceanic Engineering , 1983 , 8 ( 3 ): 173 - 184 . DOI: 10.1109/joe.1983.1145560 http://dx.doi.org/10.1109/joe.1983.1145560

Reid D . An algorithm for tracking multiple targets [J ] . IEEE Transactions on Automatic Control , 1979 , 24 ( 6 ): 843 - 854 . DOI: 10.1109/tac.1979.1102177 http://dx.doi.org/10.1109/tac.1979.1102177

Blom H A P , Bar-Shalom Y . The interacting multiple model algorithm for systems with Markovian switching coefficients [J ] . IEEE Transactions on Automatic Control , 1988 , 33 ( 8 ): 780 - 783 . DOI: 10.1109/9.1299 http://dx.doi.org/10.1109/9.1299

Li X R , Jilkov V P . Survey of maneuvering target tracking. Part Ⅴ: Multiple-model methods [J ] . IEEE Transactions on Aerospace and Electronic Systems , 2005 , 41 ( 4 ): 1255 - 1321 . DOI: 10.1109/taes.2005.1561886 http://dx.doi.org/10.1109/taes.2005.1561886

Vo B N , Ma W K . The Gaussian mixture probability hypothesis density filter [J ] . IEEE Transactions on Signal Processing , 2006 , 54 ( 11 ): 4091 - 4104 . DOI: 10.1109/tsp.2006.881190 http://dx.doi.org/10.1109/tsp.2006.881190

Reuter S , Vo B T , Vo B N , et al . The labeled multi-Bernoulli filter [J ] . IEEE Transactions on Signal Processing , 2014 , 62 ( 12 ): 3246 - 3260 . DOI: 10.1109/tsp.2014.2323064 http://dx.doi.org/10.1109/tsp.2014.2323064

杨威 , 付耀文 , 龙建乾 , 等 . 基于有限集统计学理论的目标跟踪技术研究综述 [J ] . 电子学报 , 2012 , 40 ( 7 ): 1440 - 1448 .

Yang Wei , Fu Yaowen , Long Jianqian , et al . The FISST-based target tracking techniques: A survey [J ] . Acta Electronica Sinica , 2012 , 40 ( 7 ): 1440 - 1448 . (in Chinese)

Di Kuangyu , Li Tiancheng , Li Guchong , et al . Label matching: It is complicated [C ] // 2024 27th International Conference on Information Fusion . Piscataway : IEEE , 2024 : 1 - 8 . DOI: 10.23919/fusion59988.2024.10706505 http://dx.doi.org/10.23919/fusion59988.2024.10706505

Li Wujun , Li Miao , Yi Wei , et al . Multibeam agility PRF radar multiframe decoding and track-before-detect technique [J ] . IEEE Transactions on Aerospace and Electronic Systems , 2026 , 62 : 11974 - 11991 .

Jiang Xue , Harishan K , Tharmarasa R , et al . Integrated track initialization and maintenance in heavy clutter using probabilistic data association [J ] . Signal Processing , 2014 , 94 : 241 - 250 . DOI: 10.1016/j.sigpro.2013.06.026 http://dx.doi.org/10.1016/j.sigpro.2013.06.026

Canny J . A computational approach to edge detection [J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence , 1986 , PAMI-8( 6 ): 679 - 698 . DOI: 10.1109/tpami.1986.4767851 http://dx.doi.org/10.1109/tpami.1986.4767851

Duda R O , Hart P E . Use of the Hough transformation to detect lines and curves in pictures [J ] . Communications of the ACM , 1972 , 15 ( 1 ): 11 - 15 . DOI: 10.1145/361237.361242 http://dx.doi.org/10.1145/361237.361242

Ballard D H . Generalizing the Hough transform to detect arbitrary shapes [J ] . Pattern Recognition , 1981 , 13 ( 2 ): 111 - 122 . DOI: 10.1016/0031-3203(81)90009-1 http://dx.doi.org/10.1016/0031-3203(81)90009-1

Chen Linqiang , Wang Lei , Xiong Jun . Straight-line intersection point detection based on curve fitting of Hough parameter space [C ] // 2010 International Conference on e-Education, e-Business, e-Management and e-Learning . Piscataway : IEEE , 2010 : 350 - 353 . DOI: 10.1109/ic4e.2010.65 http://dx.doi.org/10.1109/ic4e.2010.65

Fischler M A , Bolles R C . Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography [M ] // Readings in Computer Vision . Amsterdam : Elsevier , 1987 : 726 - 740 . DOI: 10.1016/b978-0-08-051581-6.50070-2 http://dx.doi.org/10.1016/b978-0-08-051581-6.50070-2

Yan Bo , Xu Na , Zhao Wenbo , et al . A three-dimensional Hough transform-based track-before-detect technique for detecting extended targets in strong clutter backgrounds [J ] . Sensors , 2019 , 19 ( 4 ): 881 . DOI: 10.3390/s19040881 http://dx.doi.org/10.3390/s19040881

Moyer L R , Spak J , Lamanna P . A multi-dimensional Hough transform-based track-before-detect technique for detecting weak targets in strong clutter backgrounds [J ] . IEEE Transactions on Aerospace and Electronic Systems , 2011 , 47 ( 4 ): 3062 - 3068 . DOI: 10.1109/taes.2011.6034689 http://dx.doi.org/10.1109/taes.2011.6034689

吴泽民 , 张磊 , 刘晗 , 等 . 基于随机Hough变换的三维集中式航迹起始算法 [J ] . 电子学报 , 2013 , 41 ( 5 ): 840 - 847 .

Wu Zemin , Zhang Lei , Liu Han , et al . Centralized 3D track initialization using random Hough transformation [J ] . Acta Electronica Sinica , 2013 , 41 ( 5 ): 840 - 847 . (in Chinese)

Rao Bin , Zhou Yongkun , Nie Yuanping . Detection and tracking of weak exoatmospheric target with elliptical Hough transform [J ] . Remote Sensing , 2022 , 14 ( 3 ): 491 . DOI: 10.3390/rs14030491 http://dx.doi.org/10.3390/rs14030491

Furgale P , Barfoot T D , Sibley G . Continuous-time batch estimation using temporal basis functions [C ] // 2012 IEEE International Conference on Robotics and Automation . Piscataway : IEEE , 2012 : 2088 - 2095 . DOI: 10.1109/icra.2012.6225005 http://dx.doi.org/10.1109/icra.2012.6225005

Hadzagic M , Michalska H . A Bayesian inference approach for batch trajectory estimation [C ] // 14th International Conference on Information Fusion . Piscataway : IEEE , 2011 : 1 - 8 .

Li Tiancheng , Chen Huimin , Sun Shudong , et al . Joint smoothing and tracking based on continuous-time target trajectory function fitting [J ] . IEEE Transactions on Automation Science and Engineering , 2019 , 16 ( 3 ): 1476 - 1483 . DOI: 10.1109/tase.2018.2882641 http://dx.doi.org/10.1109/tase.2018.2882641

Li Tiancheng , Song Yan , Fan Hongqi . From target tracking to targeting track: A data-driven yet analytical approach to joint target detection and tracking [J ] . Signal Processing , 2023 , 205 : 108883 . DOI: 10.1016/j.sigpro.2022.108883 http://dx.doi.org/10.1016/j.sigpro.2022.108883

Elsayed M N , De Silva O , Jayasiri A , et al . Optimization-based maneuvering target tracking using multiple model horizon scenario tree with model interaction [J ] . IEEE Transactions on Aerospace and Electronic Systems , 2025 , 61 ( 6 ): 15341 - 15357 . DOI: 10.1109/taes.2025.3592177 http://dx.doi.org/10.1109/taes.2025.3592177

Shi Yifang , Farina A , Song T L , et al . Distributed fusion in harsh environments using multiple bearings-only sensors with out-of-sequence-refined measurements [J ] . Aerospace Science and Technology , 2021 , 117 : 106950 . DOI: 10.1016/j.ast.2021.106950 http://dx.doi.org/10.1016/j.ast.2021.106950

Yuan Tengjiao , Zhang Qiang , Cao Heng . A highly maneuvering target tracking algorithm based on continuous-time target trajectory function [J ] . IEEE Transactions on Instrumentation and Measurement , 2026 , 75 : 1 - 12 . DOI: 10.1109/tim.2026.3655936 http://dx.doi.org/10.1109/tim.2026.3655936

Wold H . A Study in the Analysis of Stationary Time Series [M ] . Uppsala : Almqvist Wiksells Boktryckeri A.-B. , 1938 .

Li Tiancheng , Wang Jingyuan , Li Guchong , et al . From target tracking to targeting track——Part III: Stochastic process modeling and online learning [J ] . IEEE Transactions on Signal Processing , 2025 , 73 : 5336 - 5347 .

Li Tiancheng , Song Yan , Li Guchong , et al . From target tracking to targeting track-Part II: Regularized polynomial trajectory optimization [J ] . Information Fusion , 2026 , 126 : 103531 . DOI: 10.1016/j.inffus.2025.103531 http://dx.doi.org/10.1016/j.inffus.2025.103531

Rasmussen C E , Williams C K I . Gaussian Processes for Machine Learning [M ] . Cambridge, MA, USA : The MIT Press , 2005 . DOI: 10.7551/mitpress/3206.001.0001 http://dx.doi.org/10.7551/mitpress/3206.001.0001

卢春燕 , 金骁 , 邹焕新 . 层次聚类的航迹起始算法 [J ] . 信号处理 , 2013 , 29 ( 7 ): 905 - 913 .

Lu Chunyan , Jin Xiao , Zou Huanxin . Track initial algorithm based on layered clustering method [J ] . Signal Processing , 2013 , 29 ( 7 ): 905 - 913 . (in Chinese)

Chen Bing , Tugnait J K . Tracking of multiple maneuvering targets in clutter using IMM/JPDA filtering and fixed-lag smoothing [J ] . Automatica , 2001 , 37 ( 2 ): 239 - 249 . DOI: 10.1016/s0005-1098(00)00158-8 http://dx.doi.org/10.1016/s0005-1098(00)00158-8

Punchihewa Y , Vo B N , Vo B T . A generalized labeled multi-Bernoulli filter for maneuvering targets [C ] // 2016 19th International Conference on Information Fusion . Piscataway : IEEE , 2016 : 980 - 986 .

Schuhmacher D , Vo B T , Vo B N . A consistent metric for performance evaluation of multi-object filters [J ] . IEEE Transactions on Signal Processing , 2008 , 56 ( 8 ): 3447 - 3457 . DOI: 10.1109/tsp.2008.920469 http://dx.doi.org/10.1109/tsp.2008.920469

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于特征迹的密集多目标自适应视频跟踪

基于随机集的RBPF多目标关联跟踪算法

引入输入估计的交互式多模跟踪算法

一种基于H∞滤波的模糊变结构交互多模型算法

并行联合概率航迹起始逻辑