海洋测绘Issue(4)：72-77,6.DOI:10.3969/j.issn.1671-3044.2025.04.014
一种基于AIS数据的船舶习惯航线智能提取方法
An intelligent extraction method for ship habitual routes based on AIS data
摘要
Abstract
Addressing the challenges posed by the complex behavior patterns of ships,as well as the massive and high-noise data from ship AIS,an intelligent multi-disciplinary method integrating machine learning,graphics,and route design is proposed.The method builds upon ship trajectory clustering:it first constructs trajectory datasets through trajectory cleaning,stay point detection,and voyage extraction;then employs FastDTW and HDBSCAN clustering algorithms to rapidly identify ship trajectory patterns;finally utilizes the algorithm for route boundary construction,the DBA algorithm for route centerline extraction,and an improved DP algorithm for waypoint identification.Experimental results demonstrate that the proposed preprocessing method features intuitive parameter selection aligned with navigation practice,the designed trajectory clustering strategy achieves high precision and efficiency,and the extracted routes exhibit a high degree of alignment with designed routes.These findings fully validate the feasibility and effectiveness of the approach,providing a novel approach for intelligent extraction of ship habitual routes.关键词
水路运输/习惯航线提取/船舶AIS数据/轨迹预处理/轨迹聚类Key words
waterway transportation/habitual route extraction/ship AIS data/trajectory pre-processing/trajectory clustering分类
天文与地球科学引用本文复制引用
李业,王晓伟,薄锋,刘永盛,邱绍杨..一种基于AIS数据的船舶习惯航线智能提取方法[J].海洋测绘,2025,(4):72-77,6.基金项目
国家自然科学基金(52401414). （52401414）
