Enhancing vessel trajectory similarity measurement for maritime traffic monitoring using dual-component dynamic time warping

The Automatic Identification System (AIS) data provide valuable information for maritime traffic monitoring, vessel behavior analysis, and navigational safety assessment. Accurate measurement of vessel trajectory similarity is fundamental to these applications. However, most existing Dynamic Time Warping (DTW)-based methods apply a uniform alignment strategy to all trajectory points, overlooking the fact that vessel movement patterns are primarily determined by a limited number of structurally important points, while many other points mainly preserve trajectory continuity. As a result, conventional approaches may produce biased similarity measurements, particularly when trajectories exhibit similar overall movement patterns but differ in sampling density. To address this limitation, this study proposes a Dual-Component Dynamic Time Warping (DCDTW) framework for vessel trajectory similarity measurement. The proposed method decomposes vessel trajectories into feature points and non-feature points and applies differentiated similarity measurement strategies to each component according to their distinct contributions to trajectory structure and navigational behavior. By explicitly accounting for the heterogeneous importance of trajectory points, DCDTW better preserves trajectory shape characteristics and movement patterns than conventional DTW-based approaches. To support large-scale AIS data analysis, the proposed framework is further implemented within a Graphics Processing Unit (GPU)-based parallel computing environment, substantially improving computational efficiency. The performance of DCDTW is evaluated using real-world AIS trajectory datasets collected from Caofeidian Port and the Chengshan Jiao Promontory, two representative maritime regions with complex traffic conditions. Experimental results demonstrate that DCDTW consistently achieves higher trajectory similarity measurement accuracy than conventional DTW-based methods, while the GPU-accelerated implementation significantly reduces computational time for large-scale applications. The proposed framework provides an effective tool for identifying vessel behavior, analyzing maritime traffic patterns, detecting anomalies, and assessing navigational safety. By improving the reliability and scalability of trajectory similarity measurement, this study contributes to the development of data-driven approaches for maritime traffic management and safety governance in complex coastal and port environments.

Authors

Institutions

Publication Details

Journal
Ocean & Coastal Management
Published
2026-11-13
DOI
https://doi.org/10.1016/j.ocecoaman.2026.108440
Primary Topic
Maritime Navigation and Safety
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Enhancing vessel trajectory similarity measurement for maritime traffic monitoring using dual-component dynamic time warping

Hui Li, Bi Yu Chen, William Luocheng Wu, Zaili Yang et al.
2 links
Ocean & Coastal Management
Maritime Navigation and Safety
article

Enhancing vessel trajectory similarity measurement for maritime traffic monitoring using dual-component dynamic time warping

Hui Li, Bi Yu Chen, William Luocheng Wu, Zaili Yang, Yan Li, Lan Song
article en

Abstract

The Automatic Identification System (AIS) data provide valuable information for maritime traffic monitoring, vessel behavior analysis, and navigational safety assessment. Accurate measurement of vessel trajectory similarity is fundamental to these applications. However, most existing Dynamic Time Warping (DTW)-based methods apply a uniform alignment strategy to all trajectory points, overlooking the fact that vessel movement patterns are primarily determined by a limited number of structurally important points, while many other points mainly preserve trajectory continuity. As a result, conventional approaches may produce biased similarity measurements, particularly when trajectories exhibit similar overall movement patterns but differ in sampling density. To address this limitation, this study proposes a Dual-Component Dynamic Time Warping (DCDTW) framework for vessel trajectory similarity measurement. The proposed method decomposes vessel trajectories into feature points and non-feature points and applies differentiated similarity measurement strategies to each component according to their distinct contributions to trajectory structure and navigational behavior. By explicitly accounting for the heterogeneous importance of trajectory points, DCDTW better preserves trajectory shape characteristics and movement patterns than conventional DTW-based approaches. To support large-scale AIS data analysis, the proposed framework is further implemented within a Graphics Processing Unit (GPU)-based parallel computing environment, substantially improving computational efficiency. The performance of DCDTW is evaluated using real-world AIS trajectory datasets collected from Caofeidian Port and the Chengshan Jiao Promontory, two representative maritime regions with complex traffic conditions. Experimental results demonstrate that DCDTW consistently achieves higher trajectory similarity measurement accuracy than conventional DTW-based methods, while the GPU-accelerated implementation significantly reduces computational time for large-scale applications. The proposed framework provides an effective tool for identifying vessel behavior, analyzing maritime traffic patterns, detecting anomalies, and assessing navigational safety. By improving the reliability and scalability of trajectory similarity measurement, this study contributes to the development of data-driven approaches for maritime traffic management and safety governance in complex coastal and port environments.

Ocean & Coastal ManagementVol. 284
Leiden University (NL), Wuhan University (CN), State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing (CN), University of Southampton (GB), Eastern Institute of Technology, Ningbo (CN), Liverpool John Moores University (GB)
Maritime Navigation and Safety
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon