基于图谱特征的多目标航迹关联算法仿真

doi:10.16182/j.issn1004731x.joss.24-0417

摘要/Abstract

摘要：

为了解决密集场景下的多目标航迹关联问题，对多源航迹关联系统进行了建模，提出了航迹序贯实时处理的流程。通过计算目标航迹点所组成的图像中绝对位置和相对位置的图谱特征，然后基于模糊决策理论计算出统一关联矩阵，实现了目标的整体关联。仿真结果验证了基于图谱特征的航迹关联算法的有效性；特别地，在探测存在系统误差的密集目标场景下，关联的正确率要远大于最近邻距离算法，计算效率要远高于仅依赖于三角拓扑结构的关联算法，更适用于实时处理系统的应用场景。

关键词: 航迹关联, 拓扑结构, 图谱特征, 模糊决策, 序贯处理

Abstract:

In order to solve the problems of multi-targets track correlation in dense scenes, a method of track sequential real-time processing for multi-source track correlation system modelingis proposed. By calculating the absolute and relative position of the spectral features,the unified correlation matrix can be defined based on fuzzy decision theory, and the multi-target track correlation can be realized. Numerical simulations have shown the effectiveness of the track correlation algorithm on the basis of spectral features. Especially, the accuracy of correlation is much larger than that of the nearest-neighbor distance algorithm under the condition of dense target environment with systematic error. Meanwhile, the computational efficiency is much higher than that of the algorithm based on triangular topology, which proves this algorithm is more suitable for real-time processing systems.

Key words: track correlation, topology structure, spectral feature, fuzzy decision, sequential processing

中图分类号:

TP391.9

丁振平,郭徽东 . 基于图谱特征的多目标航迹关联算法仿真[J]. 系统仿真学报, 2025, 37(9): 2387-2396.

Ding Zhenping,Guo Huidong . Algorithm Simulation of Multi-targets Track Correlation Based on Spectral Feature[J]. Journal of System Simulation, 2025, 37(9): 2387-2396.

图/表 12

图1

图2

图3

表1

图4

图5

表2

图6

图7

图8

表3

图9

参考文献 16

[1]	Li Qinlei, Song Liping, Zhang Yongquan. Multiple Extended Target Tracking by Truncated JPDA in a Clutter Environment[J]. IET Signal Processing, 2021, 15(3): 207-219.
[2]	惠新成, 王珂, 蒋保富. 基于综合相似度的航迹相关方法[J]. 指挥信息系统与技术, 2014, 5(1): 37-42.
	Hui Xincheng, Wang Ke, Jiang Baofu. Track Correlation Method Based on Integrated Similarity[J]. Command Information System and Technology, 2014, 5(1): 37-42.
[3]	董凯, 王海鹏, 刘瑜. 基于拓扑统计距离的航迹抗差关联算法[J]. 电子与信息学报, 2015, 37(1): 50-55.
	Dong Kai, Wang Haipeng, Liu Yu. Anti-bias Track Association Algorithm Based on Topology Statistical Distance[J]. Journal of Electronics & Information Technology, 2015, 37(1): 50-55.
[4]	董凯, 刘瑜, 王海鹏. 基于质心参照拓扑的灰色航迹抗差关联算法[J]. 吉林大学学报(工学版), 2015, 45(4): 1311-1317.
	Dong Kai, Liu Yu, Wang Haipeng. Grey Track Anti-bias Association Algorithm Based on Centroid Topology of Reference[J]. Journal of Jilin University(Engineering and Technology Edition), 2015, 45(4): 1311-1317.
[5]	杨峰, 张妙藏, 赵露露, 等. 基于空间拓扑特征及小波变换的分布式航迹关联[J]. 指挥信息系统与技术, 2020, 11(2): 33-38.
	Yang Feng, Zhang Miaocang, Zhao Lulu, et al. Distributed Track Correlation Based on Spatial Topology Features and Wavelet Transform[J]. Command Information System and Technology, 2020, 11(2): 33-38.
[6]	毛艺帆, 张多林, 王路. 基于三角形拓扑相似性的航迹关联算法[J]. 装备学院学报, 2016, 27(4): 69-74.
	Mao Yifan, Zhang Duolin, Wang Lu. A Track Association Approach Based on TopoIogicaI TriangIe SimiIarity Degree[J]. Journal of Equipment Academy, 2016, 27(4): 69-74.
[7]	郑浩, 萨出拉. 基于拓扑结构的多元特征航迹关联算法[J]. 指挥信息系统与技术, 2020, 11(5): 83-88.
	Zheng Hao, Chula Sa. Algorithm of Multi-feature Track Association Based on Topology[J]. Command Information System and Technology, 2020, 11(5): 83-88.
[8]	崔亚奇, 熊伟, 顾祥岐. 基于三角稳定的海上目标航迹抗差关联算法[J]. 系统工程与电子技术, 2020, 42(10): 2223-2230.
	Cui Yaqi, Xiong Wei, Gu Xiangqi. Anti-bias Association Algorithm for Marine Target Track Based on Triangular Stability[J]. Systems Engineering and Electronics, 2020, 42(10): 2223-2230.
[9]	陈兆彤, 陈江平, 潘励. 基于拓扑多边形的海上船舶航迹关联方法[J]. 地球科学进展, 2022, 37(6): 588-599.
	Chen Zhaotong, Chen Jiangping, Pan Li. Correlation Method for Marine Vessel Track Based on Topological Polygon[J]. Advances in Earth Science, 2022, 37(6): 588-599.
[10]	孙乐城. 一种基于聚类的三维点模式匹配算法[D]. 长春: 吉林大学, 2020.
	Sun Lecheng. A Clustering Based Point Pattern Matching Algorithm in 3D[D]. Changchun: Jilin University, 2020.
[11]	欧鑫, 高飞, 崔浩, 等. 基于特征点匹配的点云配准方法研究[J]. 合肥工业大学学报(自然科学版), 2023, 46(6): 808-812, 821.
	Xin Ou, Gao Fei, Cui Hao, et al. Research on Point Cloud Registration Method Based on Feature Point Matching[J]. Journal of Hefei University of Technology(Natural Science), 2023, 46(6): 808-812, 821.
[12]	陈艺元. 基于图神经网络的图匹配算法研究[D]. 杭州: 杭州电子科技大学, 2022.
	Chen Yiyuan. Research on Graph Matching Algorithm Basedon Graph Neural Network[D]. Hangzhou: Hangzhou Danzi University, 2022.
[13]	陈兆彤, 陈江平, 潘励. 基于模糊隶属度的船舶中断航迹关联识别方法[J]. 应用科学学报, 2023, 41(2): 296-310.
	Chen Zhaotong, Chen Jiangping, Pan Li. Identification Method for Vessel Interrupt Track Correlating Based on Fuzzy Membership Degree[J]. Journal of Applied Sciences, 2023, 41(2): 296-310.
[14]	吴硕, 张千宇. 基于模糊多属性决策的防空武器射击决心方案选择方法[J]. 舰船电子工程, 2022, 42(12): 30-32, 93.
	Wu Shuo, Zhang Qianyu. Air Defense Weapon Fire Scheme Selection Method Based on Fuzzy Multiple Attribute Decision Making[J]. Ship Electronic Engineering, 2022, 42(12): 30-32, 93.
[15]	周万宁, 宋强, 王珂. 面向目标航迹融合的坐标变换运用技术[J]. 海军航空工程学院学报, 2018, 33(2): 205-209.
	Zhou Wanning, Song Qiang, Wang Ke. Coordinate Transformation Application Technology for Target Track Fusion[J]. Journal of Naval Aeronautical and Astronautical University, 2018, 33(2): 205-209.
[16]	董云龙, 黄高东, 李保珠. 一种非均匀系统误差的雷达分区校准方法[J]. 电光与控制, 2020, 27(3): 69-74.
	Dong Yunlong, Huang Gaodong, Li Baozhu. A Calibration Method Based on Radar Coverage Partitioning Under Non-uniform System Errors[J]. Electronics Optics & Control, 2020, 27(3): 69-74.

目标编号	雷达A探测位置/m	雷达B探测位置/m
1	(14 876.7,19 749.4)	(15 178.7,19 662.9)
2	(14 647.3,19 277.8)	(14 952.9,19 141.9)
3	(15 236.6,19 269.6)	(15 497.5,19 204.7)
4	(15 477.9,19 757.1)	(15 750.9,19 691.5)
5	(15 829.2,19 266.0)	(16 085.2,19 213.4)
6	(15 029.0,17 076.9)	(15 332.8,17 029.4)
7	(15 056.3,17 439.5)	(15 378.9,17 363.8)

综合目标	第1次关联	第2次关联	第3次关联
1	a1	a1	a1、b1
2	a2	a2	a2、b2
3	a3	a3、b2	a3、b3
4	a4	a4、b1	a4、b4
5	a5	a5、b3	a5
6	a6	a6	a6
7	a7	a7	a7
8	b1	—	—
9	b2	—	—

目标编号	航向/(°)	航向误差/%	航速/(m/s)	航速误差/%
1	180.1	6.2×10^-2	15.1	50.8×10^-2
2	180.3	17.2×10^-2	14.9	36.1×10^-2
3	180.3	19.3×10^-2	14.9	80.0×10^-2
4	180.2	10.0×10^-2	15.0	28.7×10^-2
5	180.0	0.1×10^-2	15.0	31.5×10^-2
6	238.6	57.8×10^-2	9.8	176.2×10^-2
7	305.9	3.8×10^-2	12.9	110.4×10^-2