通过对船舶AIS数据聚类可以掌握船舶运动行为和特征规律，但在轨迹聚类中通过距离描述的相似性不能连续地表征轨迹之间的相似程度，且对轨迹中的噪声点敏感、无法区分轨迹方向。针对上述问题，本文提出一种基于相似度和密度的抗噪声轨迹聚类方法，构建航向约束分段路径距离并定义轨迹相似度函数；根据轨迹相似度分布特征和聚类评价指标，建立自适应确定最佳聚类参数流程。以长江口水域AIS数据为例，基于确定的最佳参数聚类出8个不同方向的轨迹簇，结果与实际船舶习惯航路相符。实验结果表明，所提出的方法能够快速确定最佳聚类参数并对不同运动方向的轨迹进行聚类，结果可用于特征轨迹提取和航路识别，为智能航海提供技术支撑。
Ship movement behavior and characteristic laws can be grasped by clustering AIS ship data. However, the similarity described by distance in trajectory clustering cannot continuously characterize the degree of similarity between trajectories, and it is sensitive to the noise points in the trajectory and cannot distinguish the direction of the trajectory. To address the aforementioned issues, the method of ship traffic trajectory clustering based on degree of similarity and density with noise is proposed. It constructs course over ground segment-path distances and formulates trajectory similarity functions. Furthermore, a comprehensive evaluation index is defined based on the distribution characteristics of trajectory similarity. A framework for determining the optimal parameters adaptively is constructed. Finally, using AIS data from the Yangtze River estuary as a case study, this research clusters eight trajectory clusters with distinct directions based on the established optimal parameters, aligning well with real vessel navigational routes. Experimental findings underscore the method's capacity to swiftly determine optimal clustering parameters and effectively cluster trajectories across diverse motion directions, thereby facilitating feature trajectory extraction and route identification, consequently offering robust technical underpinnings for intelligent navigation.
2025,47(2): 178-184 收稿日期：2024-4-14
DOI：10.3404/j.issn.1672-7649.2025.02.029
分类号：U675
基金项目：国家社会科学基金资助项目（22BGJ034）
作者简介：杨家轩（1981 – ），男，博士，教授，研究方向为交通信息工程及控制、航海安全保障
参考文献：
[1] 刘文, 汪文博. 基于秩最小化矩阵去噪的船舶轨迹重构方法[J]. 交通运输系统工程与信息, 2022, 22(1): 106-114.
[2] RONG H, TEIXEIRA A P, SOARES C G. Maritime traffic probabilistic prediction based on ship motion pattern extraction[J]. Reliability Engineering & System safety, 2022, 217: 287-304.
[3] LIU D P, RONG H, SOARES C G. Shipping route modelling of AIS maritime traffic data at the approach to ports[J]. Ocean Engineering, 2023, 289: 120-136.
[4] ZHANG Y B, HAN Z H, ZHOU X, et al. METO-S2S: A S2S based vessel trajectory prediction method with Multiple-semantic Encoder and Type-Oriented Decoder[J]. Ocean Engineering, 2023, 277: 112-127.
[5] FARAHNAKIAN F, NICOLAS F, FARAHNAKIAN F, et al. A comprehensive study of clustering-based techniques for detecting abnormal vessel behavior[J]. Remote Sensing, 2023, 15(4): 13-37.
[6] 甄荣, 邵哲平, 潘家财, 等. 基于统计学理论的船舶轨迹异常识别[J]. 集美大学学报(自然科学版), 2015, 20(3): 193-199.
[7] ZHEN R, JIN Y X, HU Q Y, et al. Maritime anomaly detection within coastal waters based on vessel trajectory clustering and naive bayes classifier[J]. Journal of Navigation, 2017, 70(3): 648-670.
[8] YAN Z J, XIAO Y J, CHENG L, et al. Exploring AIS data for intelligent maritime routes extraction[J]. Applied Ocean Research, 2020, 101(2): 132-148.
[9] XU X F, CUI D A, LI Y, et al. Research on ship trajectory extraction based on multi-attribute dbscan optimisation algorithm[J]. Polish Maritime Research, 2021, 28(1): 136-148.
[10] 刘畅, 张仕泽, 李倍莹, 等. 考虑对地航速和航向的船舶典型轨迹提取方法[J]. 交通运输系统工程与信息, 2022, 22(3): 114-123.
[11] YANG J, LIU Y, MA L, et al. Maritime traffic flow clustering analysis by density based trajectory clustering with noise[J]. Ocean Engineering, 2022, 29: 249-262.
[12] 李倍莹, 张新宇, 沈忱, 等. 基于改进K中心点聚类的船舶典型轨迹自适应挖掘算法[J]. 上海海事大学学报, 2021, 42(3): 15-22.
[13] WEI Z, XIE X, LV W. Self-adaption vessel traffic behaviour recognition algorithm based on multi-attribute trajectory characteristics[J]. Ocean Engineering, 2020, 198: 96-110.
[14] KONTOPOULOS I, VARLAMIS I, TSERPES K. A distributed framework for extracting maritime traffic patterns[J]. International Journal of Geographical Information Science, 2021, 35(4): 767-792.
[15] SHEN H L, TANG H, YIN Y. A novel method for ship trajectory clustering[J]. International Journal of Naval Architecture and Ocean Engineering, 2022, 14: 35-49.
[16] 江玉玲, 熊振南, 唐基宏. 基于轨迹段DBSCAN的船舶轨迹聚类算法[J]. 中国航海, 2019, 42(3): 1-5.
[17] BAI X E, XIE Z X, XU X F, et al. An adaptive threshold fast DBSCAN algorithm with preserved trajectory feature points for vessel trajectory clustering[J]. Ocean Engineering, 2023, 280: 15-32.
[18] WEI Z K, GAO Y N, ZHANG X J, et al. Adaptive marine traffic behaviour pattern recognition based on multidimensional dynamic time warping and DBSCAN algorithm[J]. Expert Systems With Applications, 2024, 238: 86-102.
[19] ZHANG Y J, YUAN X Y, LI M, et al. Multi-density adaptive trajectory clustering algorithm for ships based on AIS data[J]. IEEE ACCESS, 2023, 11: 108198-108210.
[20] SHENG P, YIN J B. Extracting Shipping Route patterns by trajectory clustering model based on automatic identification system data[J]. Sustainability, 2018, 10: 71-87.
[21] PENG X, FENG J S, XIAO S J, et al. Structured autoencoders for subspace clustering[J]. IEEE Transactions on Image Processing, 2018, 27: 5076-5086.
