基于MLS的三维扩展目标PMBM跟踪算法

doi:10.12305/j.issn.1001-506X.2023.11.06

摘要/Abstract

摘要：

针对传统三维扩展目标跟踪算法形状估计精度低的问题, 提出了一种基于移动最小二乘的泊松多伯努利混合(Poisson multi-Bernoulli mixture based on the moving least square, MLS-PMBM)滤波跟踪算法。该算法基于MLS模型构建三维扩展目标的形状矩阵, 通过PMBM滤波器预测和更新目标的运动状态, 利用移动最小二乘算法更新形状矩阵, 结合目标质心状态与形状估计完成对三维扩展目标的跟踪。仿真实验与实际点云数据的验证表明, 与现有算法相比, 本文所提算法在多扩展目标的形状估计方面具有更优的性能, 具有较高的泛用性。

关键词: 多扩展目标跟踪, 移动最小二乘, 形状矩阵, 泊松多伯努利混合

Abstract:

Aiming at the problem of low accuracy of shape estimation in traditional 3D extended target tracking algorithms, a of filtering tracking algorithm Poisson multi-Bernoulli mixture based on the moving least square (MLS-PMBM) is proposed. The algorithm constructs the shape matrix of the 3D extended target based on the MLS model, predicted and updated the motion state of the target by PMBM filter. The shape matrix is updated by the moving least squares algorithm, and the tracking of the 3D extended target is completed by combining the centroid state and shape estimation of the target. Simulation experiments and actual point cloud data verification show that compared with the existing algorithms, the proposed algorithm has better performance in the shape estimation of multiple extended objects, and has higher universality.

Key words: multiple extended target tracking, moving least square (MLS), shape matrix, Poisson multi-Bernoulli mixture (PMBM)

中图分类号:

TN953

衡博文, 李翠芸, 李想. 基于MLS的三维扩展目标PMBM跟踪算法[J]. 系统工程与电子技术, 2023, 45(11): 3411-3418.

Bowen HENG, Cuiyun LI, Xiang LI. 3D extended target PMBM tracking algorithm based on moving least square[J]. Systems Engineering and Electronics, 2023, 45(11): 3411-3418.

图/表 20

图1

图2

图3

表1

图4

图5

图6

图7

图8

图9

表2

表3

图10

图11

图12

表4

图13

图14

图15

图16

参考文献 22

1	MAHLER R . Statistical multisource-multitarget information fusion[M]. Norwood: Artech House, 2007.
2	WILLIAMS J L . Marginal multi-Bernoulli filters: RFS derivation of MHT, JIPDA, and association-based MeMBer[J]. IEEE Trans.on Aerospace and Electronic Systems, 2015, 51 (3): 1664- 1687. doi: 10.1109/TAES.2015.130550
3	GARCÍA-FERNÁNDEZ Á F , WILLIAMS J L , GRANSTRÖM K , et al. Poisson multi-Bernoulli mixture filter: direct derivation and implementation[J]. IEEE Trans.on Aerospace and Electronic Systems, 2018, 54 (4): 1883- 1901. doi: 10.1109/TAES.2018.2805153
4	KOCH J W . Bayesian approach to extended object and cluster tracking using random matrices[J]. IEEE Trans.on Aerospace and Electronic Systems, 2008, 44 (3): 1042- 1059. doi: 10.1109/TAES.2008.4655362
5	BAUM M, HANEBECK U D. Random hypersurface models for extended object tracking[C]//Proc. of the IEEE International Symposium on Signal Processing and Information Technology, 2009: 178-183.
6	DANIYAN A , LAMBOTHARAN S , DELIGIANNIS A , et al. Bayesian multiple extended target tracking using labeled random finite sets and splines[J]. IEEE Trans.on Signal Processing, 2018, 66 (22): 6076- 6091. doi: 10.1109/TSP.2018.2873537
7	DAHLÉN K M, LINDBERG C, YONEDA M, et al. An improved B-spline extended object tracking model using the iterative closest point method[C]//Proc. of the IEEE 25th International Conference on Information Fusion, 2022.
8	HIRSCHER T, SCHEEL A, REUTER S, et al. Multiple extended object tracking using Gaussian processes[C]//Proc. of the IEEE 19th International Conference on Information Fusion, 2016: 868-875.
9	陈辉, 杜金瑞, 韩崇昭. 基于星凸形随机超曲面模型多扩展目标多伯努利滤波器[J]. 自动化学报, 2020, 46 (5): 909- 922.
	CHEN H , DU J R , HAN C Z . A multiple extended target multi-Bernouli filter based on star-convex random hypersurface model[J]. Acta Automatica Sinica, 2020, 46 (5): 909- 922.
10	张永权, 张海涛, 姬红兵. 基于椭圆RHM的扩展目标伯努利滤波算法[J]. 系统工程与电子技术, 2018, 40 (9): 1905- 1910. doi: 10.3969/j.issn.1001-506X.2018.09.02
	ZHANG Y Q , ZHANG H T , JI H B . Extended-target Bernoulli filter algorithm based on ellipse RHM[J]. Systems Engineering and Electronics, 2018, 40 (9): 1905- 1910. doi: 10.3969/j.issn.1001-506X.2018.09.02
11	LAN J , LI X R . Tracking of maneuvering non-ellipsoidal extended object or target group using random matrix[J]. IEEE Trans.on Signal Processing, 2014, 62 (9): 2450- 2463. doi: 10.1109/TSP.2014.2309561
12	GRANSTRÖM K, WILLETT P, BAR-SHALOM Y. An extended target tracking model with multiple random matrices and unified kinematics[C]//Proc. of the IEEE 18th International Conference on Information Fusion, 2015: 1007-1014.
13	LAN J, LI X R. Tracking of extended object or target group using random matrix—Part Ⅱ: irregular object[C]//Proc. of the IEEE 15th International Conference on Information Fusion, 2012: 2185-2192.
14	GONG Y , CUI C , WU B . A GGIW-PHD filter for multiple non-ellipsoidal extended targets tracking with varying number of sub-objects[J]. IEEE Access, 2021, 9, 64719- 64731. doi: 10.1109/ACCESS.2021.3075941
15	范泽昊. 基于泊松多伯努利混合滤波器的三维多扩展目标跟踪算法研究[D]. 西安: 西安电子科技大学, 2021: 25-45.
	FAN Z H. Research on 3D multiple extended target tracking algorithm based on Poisson multi-Bernoulli mixture filter[D]. Xi'an: Xidian University, 2021: 25-45.
16	GRANSTRÖM K, FATEMI M, SVENSSON L. Gamma Gaussian inverse-Wishart Poisson multi-Bernoulli filter for extended target tracking[C]//Proc. of the IEEE 19th International Conference on Information Fusion, 2016: 893-900.
17	FAION F, BAUM M, HANEBECK U D. Tracking 3D shapes in noisy point clouds with random hypersurface models[C]//Proc. of the IEEE 15th International Conference on Information Fusion, 2012: 2230-2235.
18	EBERT F, WUENSCHE H J. Dynamic object tracking and 3D surface estimation using Gaussian processes and extended Kalman filter[C]//Proc. of the IEEE Intelligent Transportation Systems Conference, 2019: 1122-1127.
19	SCHAEFER S , MCPHAIL T , WARREN J . Image deformation using moving least squares[J]. ACM Trans.on Graphics, 2006, 25 (3): 533- 540. doi: 10.1145/1141911.1141920
20	LEVIN D . The approximation power of moving least-squares[J]. Mathematics of Computation, 1998, 67 (224): 1517- 1531. doi: 10.1090/S0025-5718-98-00974-0
21	CAO X, LAN J, LI X R. Extension-deformation approach to extended object tracking[C]//Proc. of the IEEE 19th International Conference on Information Fusion, 2016: 1185-1192.
22	ZHOU D, FANG J, SONG X, et al. Iou loss for 2D/3D object detection[C]//Proc. of the IEEE International Conference on 3D Vision (3DV), 2019: 85-94.

目标	目标初始运动状态	存活时间/s
1	x₁=[－175, －175, －175, 7, 7, 7]^T	t₁∈[1, 50]
2	x₂=[－175, 150, 120, 5, －5, －6]^T	t₂∈[1, 30]
3	x₃=[－175, －150, 140, －5, 7, －5]^T	t₃∈[20, 50]

目标类型	算法
目标类型	MLS-PMBM	GGIW-PMBM
椭球	0.912 3	0.909 2
立方体	0.841 9	0.217 4
十字架	0.823 3	0.243 8

算法	估计椭球目标形状时算法的单步耗时	估计立方体目标形状时算法的单步耗时	估计十字架目标形状时算法的单步耗时
MLS-PMBM	0.631 0	0.652 1	0.621 2
GGIW-PMBM	0.263 6	0.249 3	0.245 2

目标	目标运动状态	时间/s
1	x₁=[0, 0, 126, 7, 0, 0]^T	t₁∈[1, 30]
2	x₂=[38, 2, 126, 10, 0, 0]^T	t₂∈[9, 15]
	x₂=[98, 2, 126, 7, 0, 0]^T	t₂∈[15, 18]
	x₂=[119, 2, 126, 20, 0, 0]^T	t₂∈[18, 30]

[1]	毕文豪, 周杰, 张安, 刘力. 杂波环境下基于最大熵模糊聚类的JPDA算法[J]. 系统工程与电子技术, 2023, 45(7): 1920-1927.
[2]	王森, 鲍庆龙, 潘嘉蒙, 祝茜. 基于改进概率假设密度滤波器的非合作双基地雷达目标跟踪[J]. 系统工程与电子技术, 2023, 45(7): 2002-2009.
[3]	赵石磊, 陈万春, 杨良, 冯卓. 大气层内助推段广义标控脱靶量解析制导律[J]. 系统工程与电子技术, 2023, 45(7): 2150-2157.
[4]	张宏伟. 双站纯方位空时软约束无迹粒子滤波算法[J]. 系统工程与电子技术, 2023, 45(5): 1261-1269.
[5]	任明健, 胡国平, 周豪, 游致远, 张凌培. 基于耦合张量分解的稀疏阵列二维DOA估计算法[J]. 系统工程与电子技术, 2023, 45(4): 958-964.
[6]	刘政玮, 陈映, 鲁耀兵. 适用于多目标轨迹小角度交叉的PHD滤波器[J]. 系统工程与电子技术, 2023, 45(4): 982-990.
[7]	张育豪, 朱圣棋, 曾操, 崔森, 石琦剑. EPC-MIMO雷达主瓣距离欺骗式干扰抑制方法[J]. 系统工程与电子技术, 2023, 45(3): 690-698.
[8]	游致远, 胡国平, 周豪. 基于冗余阵元优化的双基地嵌套MIMO雷达目标DOD和DOA联合估计方法[J]. 系统工程与电子技术, 2022, 44(12): 3696-3702.
[9]	张瑜, 吴凯, 郭杰, 葛志闪, 张宝超. 基于数据质量评估的自适应序贯航迹关联算法[J]. 系统工程与电子技术, 2022, 44(11): 3477-3485.
[10]	刘浩楠, 宋骊平. 基于核Fisher判别的群结构更新模型及群目标跟踪算法[J]. 系统工程与电子技术, 2022, 44(10): 3012-3019.
[11]	孙照强, 王志贵, 孟飞, 李陆雨, 于中, 陈燕. 基于EKF及弹道方程的弹道目标跟踪滤波器设计[J]. 系统工程与电子技术, 2022, 44(10): 3207-3212.
[12]	万福海, 许京伟, 张振荣. FDA-MIMO雷达稳健抗主瓣距离欺骗式干扰技术[J]. 系统工程与电子技术, 2022, 44(9): 2809-2816.
[13]	侯子林, 程婷, 彭瀚. 基于量测转换序贯滤波的GMPHD机动目标跟踪[J]. 系统工程与电子技术, 2022, 44(8): 2474-2482.
[14]	刘红亮, 陈超, 黄昆伟, 卢博, 岳凯. 针对分布式雷达网的多帧联合航迹起始方法[J]. 系统工程与电子技术, 2022, 44(7): 2143-2147.
[15]	王伟, 王钦钊, 沈岚, 王波, 丁怀. 基于概率累积的非通视无源定位方法[J]. 系统工程与电子技术, 2022, 44(1): 64-69.