在目标跟踪领域,复杂机动目标的动态特性对算法的鲁棒性与精度提出严峻挑战。传统交互式多模型算法由于采用固定模型参数与转移概率矩阵,在运动模型切换时易产生响应滞后与较大瞬态误差;同时,对输入序列时序特征的挖掘能力有限,导致复杂场景下跟踪性能受限。为此,本文在BiLSTM-IMM框架上提出DCGLA-IMM算法,通过深度时序建模与注意力机制增强特征提取和动态适应能力。算法利用可学习膨胀率的因果卷积实现多时间尺度局部特征提取,结合残差连接的双层BiLSTM强化长程时序建模,并引入全局—局部门控混合注意力以实现局部细节与全局趋势的自适应平衡;最后由GRU整合序列信息,通过多预测头输出各运动模型概率,并完成概率校正与状态融合。仿真实验结果表明,在同结构运动模型参数时变与多模型切换2类复杂场景下,所提算法相较BiLSTM-IMM的RMSE分别降低10.07%与9.14%,在模型概率响应延迟与概率质量等指标上亦取得更优表现,验证了该方法在复杂机动目标跟踪中的精度提升与鲁棒性增强。
In target tracking, the dynamic characteristics of highly maneuvering targets pose significant challenges to the robustness and accuracy of tracking algorithms. Traditional Interactive Multiple Model algorithms employ fixed model parameters and transition probability matrices, which often lead to response delays and large transient errors during motion model switching. In addition, their limited capability in exploiting temporal features of input sequences restricts tracking performance in complex scenarios. To address these issues, this paper proposes a DCGLA-IMM algorithm based on the BiLSTM-IMM framework, in which deep temporal modeling and attention mechanisms are introduced to enhance feature extraction and dynamic adaptability. Specifically, causal convolution with learnable dilation rates is utilized to extract multi-scale local temporal features, while a two-layer BiLSTM with residual connections is employed to strengthen long-range temporal modeling. Furthermore, a global-local gated hybrid attention mechanism is incorporated to achieve adaptive fusion of local details and global trends. Finally, a GRU integrates sequential information, and multiple prediction heads output the probabilities of different motion models, followed by probability correction and state fusion. Simulation results demonstrate that, under two challenging scenarios with time-varying parameters of fixed-structure motion models and multi-model switching, the proposed algorithm reduces the RMSE by 10.07% and 9.14%, respectively, compared with the BiLSTM-IMM. It also achieves superior performance in terms of model probability response delay and probability quality metrics, thereby validating its improved accuracy and robustness for tracking highly maneuvering targets.
[1] Ying Z Y, Li H.IMM-SLAMMOT: Tightly-coupled SLAM and IMM-based multi-object tracking[J]. IEEE Transactions on Intelligent Vehicles, 2024, 9(2): 3964-3974. DOI:10.1109/TIV.2023.3346040.
[2] Pei J D, Wang Y D, Deng S R, et al.Space-based maneuvering target tracking algorithm based on IMM[C]//Advances in Guidance, Navigation and Control. Singapore: Springer, 2025: 157-167. DOI:10.1007/978-981-96-2260-3_16.
[3] Singer R A. Estimating optimal tracking filter performance for manned maneuvering targets[J]. IEEE Transactions on Aerospace and Electronic Systems, 1970, AES-6(4): 473-483. DOI:10.1109/TAES.1970.310128.
[4] Zhao Z W, Chen H.Multi-maneuvering target tracking based on a Gaussian process[J]. Sensors, 2024, 24(22): 7270. DOI:10.3390/s24227270
[5] Dong S L, Liu M Q, Wu Z G.A survey on hidden Markov jump systems: Asynchronous control and filtering[J]. International Journal of Systems Science, 2023, 54(6): 1360-1376. DOI:10.1080/00207721.2023.2171710.
[6] 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.
[7] 曾浩,母王强,蒋阳,等. 模型参数自适应的低复杂度 ATPM-VSIMM 算法[J]. 通信学报,2023, 44(9): 25-35. DOI:10.11959/j.issn.1000-436x.2023186.
[8] 汪晋, 苏洪涛, 汪圣利, 等. 基于SLSTM网络的两级修正机动目标跟踪方法[J]. 西安电子科技大学学报, 2025, 52(1): 37-49. DOI:10.19665/j.issn1001-2400.20241015.
[9] Hang G, Wang J W.Adaptive modified transition probability IMM algorithm for maneuvering target tracking[C]//2024 9th International Conference on Electronic Technology and Information Science (ICETIS). May 17-19, 2024, Hangzhou, China. IEEE, 2024: 427-431. DOI:10.1109/ICETIS61828.2024.10593671.
[10] 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.
[11] Lee I H, Park C G.A two-stage transition correction function for adaptive Markov matrix in IMM algorithm[C]//2022 25th International Conference on Information Fusion (FUSION). July 4-7, 2022, Linköping, Sweden. IEEE, 2022: 1-8. DOI:10.23919/FUSION49751.2022.9841371.
[12] 赵申奇. 基于卡尔曼滤波和神经网络的水下目标跟踪研究[D]. 桂林: 桂林电子科技大学, 2024.
[13] Elkateb S, Métwalli A, Shendy A, et al.Machine learning and IoT-Based predictive maintenance approach for industrial applications[J]. Alexandria Engineering Journal, 2024, 88: 298-309. DOI:10.1016/j.aej.2023.12.065.
[14] Kailash Varma N M, Ahmed M I, Madhusudhan G, et al. Machine learning-enabled smart transit: Real-time bus tracking system for enhanced urban mobility[C]//2024 Intelligent Systems and Machine Learning Conference (ISML). May 4-5, 2024, Hyderabad, India. IEEE, 2025: 520-524. DOI:10.1109/ISML60050.2024.11007439.
[15] Malashin I, Tynchenko V, Gantimurov A, et al.Applications of long short-term memory (LSTM) networks in polymeric sciences: A review[J]. Polymers, 2024, 16(18): 2607. DOI:10.3390/polym16182607.
[16] Aftab W, Mihaylova L.A learning Gaussian process approach for maneuvering target tracking and smoothing[J]. IEEE Transactions on Aerospace and Electronic Systems, 2021, 57(1): 278-292. DOI:10.1109/TAES.2020.3021220.
[17] Zhang L, Lei Z C, Le Y J, et al.Online recursive Gaussian process for target location based on ultra-short baselines[C]//2024 36th Chinese Control and Decision Conference (CCDC). May 25-27, 2024, Xi’an, China. IEEE, 2024: 4361-4366. DOI:10.1109/CCDC62350.2024.10587757.
[18] Shen-tu H, Zhang H Y, Zhu Y M, et al. A variable structure multi-model maneuvering target tracking algorithm based on Monte Carlo learning[J]. IET Radar, Sonar & Navigation, 2023, 17(12): 1785-1795. DOI:10.1049/rsn2.12464.
[19] 董浩. 移动传感器网络中复杂运动轨迹目标的智能跟踪调度算法研究[D]. 成都: 电子科技大学, 2022.
[20] Xia L W, Zhang Y, Liu H J.Online optimization and feedback Elman neural network for maneuvering target tracking[C]//2017 4th IAPR Asian Conference on Pattern Recognition (ACPR). November 26-29, 2017, Nanjing, China. IEEE, 2018: 494-499. DOI:10.1109/ACPR.2017.56.
[21] Zhu Y A, Jia Z Y, Wu Q H, et al.UAV trajectory tracking via RNN-enhanced IMM-KF with ADS-B data[C]//2024 IEEE Wireless Communications and Networking Conference (WCNC). April 21-24, 2024, Dubai, United Arab Emirates. IEEE, 2024: 1-6. DOI:10.1109/WCNC57260.2024.10570914.
[22] Yu W T, Yu H Y, Du J P, et al.DeepGTT: A general trajectory tracking deep learning algorithm based on dynamic law learning[J]. IET Radar, Sonar & Navigation, 2021, 15(9): 1125-1150. DOI:10.1049/rsn2.12111.
[23] Moon S, Youn W, Bang H.Novel deep-learning-aided multimodal target tracking[J]. IEEE Sensors Journal, 2021, 21(18): 20730-20739. DOI:10.1109/JSEN.2021.3100588.
[24] 朱一鸣. 结合时序神经网络的多模型机动目标跟踪算法研究[D]. 杭州: 杭州电子科技大学, 2024.
