自适应IMM-UKF机动目标跟踪算法

doi:10.12305/j.issn.1001-506X.2025.08.26

摘要/Abstract

摘要：

针对跟踪复杂机动目标过程中由于目标运动状态发生变化导致的跟踪误差较大的问题，提出一种自适应交互多模型无迹卡尔曼滤波（interacting multiple model unscented Kalman filter，IMM-UKF）算法，使用模型概率后验信息和模型似然函数自适应修正马尔可夫转移概率矩阵（transition probability matrix，TPM）。设计模型概率校正方法和模型转移加速方法，两种方法分别作用于模型稳定阶段和模型转移阶段，提高模型概率准确度和模型转移响应速度，减小状态估计误差。最后，通过两种场景下的实验验证所提算法在目标具有复杂运动状态下的性能，并与传统方法进行对比分析，在目标做机动运动时，位置精度和速度精度分别提高了15%和26%，验证了算法的有效性和可行性。

关键词: 目标跟踪, 交互多模型, 自适应, 无迹卡尔曼滤波

Abstract:

Aiming at the problem of large tracking error caused by the change of target motion state in the process of tracking complex maneuvering targets, an adaptive interacting multiple model unscented kalman filter （IMM-UKF） algorithm is proposed, which uses model probability posterior information and model likelihood function to adaptively modify Markov transition probability matrix （TPM）. The model probability correction method and model transition acceleration method are designed. The two methods act on the model stability stage and model transition stage respectively, which improve the model probability accuracy and model transition response speed, and reduce the state estimation error. Finally, the performance of the proposed algorithm in the target with complex motion is verified by experiments in two scenarios, and compared with the traditional methods, the position accuracy and velocity accuracy are improved by 15% and 26% respectively when the target is maneuvering, which verifies the effectiveness and feasibility of the algorithm.

Key words: target tracking, interacting multiple model （IMM）, adaptive, unscented Kalman filter （UKF）

中图分类号:

TN 953

周晓, 牟新刚, 柯文, 苏盈, 王丽. 自适应IMM-UKF机动目标跟踪算法[J]. 系统工程与电子技术, 2025, 47(8): 2686-2695.

Xiao ZHOU, Xingang MOU, Wen KE, Ying SU, Li WANG. Adaptive IMM-UKF maneuvering target tracking algorithm[J]. Systems Engineering and Electronics, 2025, 47(8): 2686-2695.

图/表 17

图1

图2

图3

表1

图4

图5

表2

图6

图7

图8

图9

图10

图11

图12

图13

图14

表3

参考文献 31

1	WANG J, HE J C, PENG B, et al. Generalized interacting multiple model Kalman filtering algorithm for maneuvering target tracking under non-Gaussian noises[J]. ISA Transaction, 2024, 155, 148- 163. doi: 10.1016/j.isatra.2024.09.015
2	程雨. 基于单测向站的目标跟踪算法研究[D]. 武汉: 华中科技大学, 2019.
	CHENG Y. Research on target tracking algorithm based on single direction finding station[D]. Wuhan: Huazhong University of Science and Technology, 2019.
3	王芝燕. 低空机动目标跟踪算法研究[D]. 成都: 电子科技大学, 2022.
	WANG Z Y. Research on low altitude maneuvering target tracking algorithm[D]. Chengdu: University of Electronic Science and Technology of China, 2022.
4	YANG B J, WANG H G, SHI Z Y. Interacting multiple model adaptive robust Kalman filter for process and measurement modeling errors simultaneously[J]. Signal Processing, 2024, 227, 109743.
5	ARROYO C A, ASENSIO V M. Adaptive IMM-UKF for airborne tracking[J]. Aerospace, 2023, 10 (8): 698. doi: 10.3390/aerospace10080698
6	XIE G, SUN L L, WEN T, et al. Adaptive transition probability matrix-based parallel IMM algorithm[J]. IEEE Trans. on Systems, Man, and Cybernetics: Systems, 2021, 51 (5): 2980- 2989.
7	XU W W, XIAO J K, XU D L, et al. An adaptive IMM algorithm for a PD radar with improved maneuvering target tracking performance[J]. Remote Sensing, 2024, 16 (6): 1051. doi: 10.3390/rs16061051
8	王平波, 陈强, 卫红凯, 等. 一种基于模型概率单调性变化的自适应IMM-UKF改进算法[J]. 电子与信息学报, 2024, 46 (1): 41- 48.
	WANG P B, CHEN Q, WEI H K, et al. An improved adaptive IMM-UKF algorithm based on monotonicity change of model probability[J]. Journal of Electronics & Information Technology, 2024, 46 (1): 41- 48.
9	臧荣春, 崔平远. 马尔可夫参数自适应IFIMM算法研究[J]. 电子学报, 2006, 34 (3): 521- 524.
	ZANG R C, CUI P Y. Research on adaptive Markov parameter IFIMM algorithm[J]. Acta Electronica Sinica, 2006, 34 (3): 521- 524.
10	封普文, 黄长强, 曹林平, 等. 马尔可夫矩阵修正IMM跟踪算法[J]. 系统工程与电子技术, 2013, 35 (11): 2269- 2274.
	FENG P W, HUANG C Q, CAO L P, et al. Markov matrix modified IMM tracking algorithm[J]. Systems Engineering and Electronics, 2013, 35 (11): 2269- 2274.
11	LEE I H, PARK C G. An improved interacting multiple model algorithm with adaptive transition probability matrix based on the situation[J]. International Journal of Control, Automation and Systems, 2023, 21 (10): 3299- 3312.
12	DENG L C, LI D, LI R F. Improved IMM algorithm based on RNNs[J]. Journal of Physics: Conference Series, 2020, 1518 (1): 12055. doi: 10.1088/1742-6596/1518/1/012055
13	BECKER S, HUG R, HUEBNER W, et al. An RNN-based IMM filter surrogate[J]. Image Analysis, 2019, 11482, 387- 398.
14	LUI D G, TARTAGLIONE G, CONTI F, et al. Long short-term memory-based neural networks for missile maneuvers trajectories prediction[J]. IEEE Access, 2023, 11, 30819- 30831. doi: 10.1109/ACCESS.2023.3262023
15	LIU B C, SHI Q Y, HAN P. Short-term 4D trajectory prediction method based on LST-M-IMM[C]//Proc. of the IEEE/AIAA 41st Digital Avionics Systems Conference, 2022.
16	SUN Y, TIAN X, BAO W, et al. Improving treatment of noise specification of Kalman filtering for state updating of hydrological models: combining the strengths of the interacting multiple model method and c-ubature Kalman filter[J]. Water Resources Research, 2023, 59 (7): e2022WR033635. doi: 10.1029/2022WR033635
17	DENG F, YANG H L, WANG L J. Adaptive unscented Kalman filter based estimation and filtering for dynamic positioning with model uncertainties[J]. International Journal of Control, Automation and Systems, 2019, 17 (3): 667- 678.
18	HU K Y, WANG J M, CHENG Y Q, et al. Adaptive filtering and smoothing algorithm based on variable structure interactive multiple model[J]. Scientific Reports, 2023, 13 (1): 12993. doi: 10.1038/s41598-023-39075-9
19	ZHOU H, ZHAO H, HUANG H Q, et al. A cubature-principle-assisted IMM-adaptive UKF algorithm for maneuvering target tracking caused by sensor faults[J]. Applied Sciences, 2017, 7 (10): 1003. doi: 10.3390/app7101003
20	WU X T, LIU Y, MA X C. An underwater maneuvering target tracking algorithm based on UKF with adaptive sampling range[C]//Proc. of the Technical Committee on Control Theory, 2023: 3133−3138.
21	LIU M, NIU J, LIU Y. UKF-MOT: an unscented Kalman filter-based 3D multi-object tracker[J]. CAAI Transactions on Intelligence Technology, 2024, 9 (4): 1031- 1041. doi: 10.1049/cit2.12315
22	DONG Y L, LI W Q, LI D X, et al. Intelligent tracking method for aerial maneuvering target based on unscented Kalman filter[J]. Remote Sensing, 2024, 16 (17): 3301. doi: 10.3390/rs16173301
23	RONG D D, WANG Y. An adaptive spatial target tracking method based on unscented Kalman filter[J]. Sensors, 2024, 24 (18): 6094. doi: 10.3390/s24186094
24	HOU Z W, BU F L. A small UAV tracking algorithm based on AIMM-UKF[J]. Aircraft Engineering and Aerospace Technology, 2021, 93 (4): 579- 591. doi: 10.1108/AEAT-01-2019-0013
25	黄小平, 王岩, 缪鹏程. 目标定位跟踪原理及应用[M]. 北京: 电子工业出版社, 2018.
	HUANG X P, WANG Y, MIAO P C. Principle and application of target location and tracking[M]. Beijing: Publishing House of Electronics Industry, 2018.
26	ZHAO B F. Multisensor maneuvering target fusion tracking using interacting multiple model[J]. Automatic Control and Computer Sciences, 2024, 58 (3): 303- 312. doi: 10.3103/S0146411624700184
27	SUN M X, DUAN Q W, XIA W R, et al. Multiple adaptive factors based interacting multiple model estimator[J]. IET Control Theory & Applications, 2024, 18 (8): 1059- 1069.
28	LI X H, LU B, LI Y X, et al. Adaptive interacting multiple model for underwater maneuvering target tracking with one-step randomly delayed measurements[J]. Ocean Engineering, 2023, 280, 114933. doi: 10.1016/j.oceaneng.2023.114933
29	ZHENG T Y, YAO Y, HE F H, et al. Active switching multiple model method for tracking a noncooperative gliding flight vehicle[J]. SCIENCE CHINA Information Sciences, 2020, 63 (9): 192202. doi: 10.1007/s11432-019-1515-2
30	WANG J W, CHEN X Y, SHAO X. An adaptive multiple backtracking UKF method based on Krein space theory for marine vehicles alignment process[J]. IEEE Trans. on Vehicular Technology, 2023, 72 (3): 3214- 3226. doi: 10.1109/TVT.2022.3220243
31	QIU Y H, SU Y, HE X Y, et al. An improved IMM-KF for UAV position prediction[C]//Proc. of the International Conference on Control and Intelligent Robotics, 2023.

运动模型	时间/s	运动参数
CA	1~200	a = 10 m/s²
CV	201~400	v = 300 m/s
CT	401~600	ω = 0.02 rad/s

算法	位置/m	速度/（m/s）
IMM-KF	82.15	25.22
IMM-UKF	78.55	25.17
AIMM-UKF	76.89	16.55
ATPM-UKF	65.75	12.17

算法	位置/m	速度/（m/s）
IMM-KF	10.81	14.21
IMM-UKF	9.43	11.74
AIMM-UKF	9.40	11.47
ATPM-UKF	8.13	10.26

[1]	马晓萌, 邓东明, 沈永健, 丁金闪, 郝国卿. 基于单站无源运动定位的多目标跟踪方法[J]. 系统工程与电子技术, 2025, 47(8): 2549-2557.
[2]	姬博洋, 李炯, 贺杨超, 叶继坤, 雷虎民. 基于最小代价准则的再入滑翔制导方法[J]. 系统工程与电子技术, 2025, 47(8): 2715-2726.
[3]	李晓辉, 杨雯, 吕思婷, 毛亮. 基于决策树的多视觉任务自适应压缩算法[J]. 系统工程与电子技术, 2025, 47(8): 2737-2743.
[4]	孙一心, 方愚渊, 张磊. 战术机动先验辅助的交互多模型目标跟踪方法[J]. 系统工程与电子技术, 2025, 47(7): 2086-2097.
[5]	马治勋, 殷超然, 王天琪, 郝程鹏. 融入先验知识的MIMO声呐自适应检测方法[J]. 系统工程与电子技术, 2025, 47(7): 2383-2388.
[6]	赵正义, 南普龙. 基于短时傅里叶变换的高效空频自适应抗干扰方法[J]. 系统工程与电子技术, 2025, 47(6): 1778-1785.
[7]	张洪源, 龚柏春, 韩飞, 孙玥, 宁昕. 复杂环境下天基无源协同多目标初始定轨方法[J]. 系统工程与电子技术, 2025, 47(5): 1404-1413.
[8]	闫咏琪, 王琪, 易凡. 交互式多模型概率数据关联抗速度拖引干扰算法[J]. 系统工程与电子技术, 2025, 47(5): 1507-1515.
[9]	向怡然, 李松亭, 陈利虎. 基于聚类算法的星载ADS-B波束自适应调整方法[J]. 系统工程与电子技术, 2025, 47(5): 1680-1686.
[10]	陈威, 王从庆, 曾强, 李战. 面向飞机表面视觉检查的无人机覆盖路径规划[J]. 系统工程与电子技术, 2025, 47(4): 1206-1213.
[11]	邢胜杰, 赵冬, 任文静. 基于事件触发的柔性机械臂执行器故障补偿[J]. 系统工程与电子技术, 2025, 47(4): 1311-1318.
[12]	于少筠, 谢菊兰, 何子述, 李会勇. 基于整数时延的宽带LFM信号最优自适应波束形成[J]. 系统工程与电子技术, 2025, 47(3): 691-700.
[13]	龙锦荣, 吴建新, 梁毅. 分布式相参雷达自适应和差波束测角方法[J]. 系统工程与电子技术, 2025, 47(3): 779-787.
[14]	黄绍洧, 都延丽, 刘燕斌, 王跃萍, 刘武. 有限时间收敛的自适应滑模协同末制导[J]. 系统工程与电子技术, 2025, 47(3): 961-969.
[15]	宋怡佳, 王海燕, 冯伟, 全英汇. 基于高光谱跨场景迁移学习的集成对齐子空间自适应算法[J]. 系统工程与电子技术, 2025, 47(2): 369-375.