考虑交通流特征的船舶轨迹相似性度量方法  

A Ship Trajectory Clustering Method Considering the Spatiotemporal Characteristics of Traffic Flow

作　　者：李烊 齐乐 梁高成 胡银杰 季园园 Robert Balling LI Yang;QI Le;LIANG Gao-cheng;HU Yin-jie;JI Yuan-yuan;Robert Balling(School of Navigation,Wuhan University of Technology,Wuhan 430063,China;National Engineering Research Center for Water Transport Safety,Wuhan University of Technology,Wuhan 430063,China;State Key Laboratory of Martime Technology and Safety,Wuhan University of Technology,Wuhan 430063,China;School of Geographical Sciences and Urban Planning,Arizona State University,Tempe 85287,America)

机构地区：[1]武汉理工大学航运学院,武汉430063 [2]武汉理工大学国家水运安全工程技术研究中心,武汉430063 [3]武汉理工大学水路交通控制全国重点实验室,武汉430063 [4]亚利桑那州立大学地理科学与城市规划学院,坦佩85287

出　　处：《武汉理工大学学报》2024年第7期107-115,共9页Journal of Wuhan University of Technology

基　　金：湖北省自然科学基金(2019CFB339);国家科技部,国家重点研发计划课题-子任务(2023YFB4302304)。

摘　　要：近年来数据挖掘技术被广泛应用于水上船舶交通研究方面,基于AIS获取的大量船舶轨迹数据中含有船舶行为特征,如何通过这些行为特征衡量船舶轨迹间的相似性,提取具有相似运动规律的船舶轨迹是当前研究的重点与热点问题。作者从船舶交通流的特征出发,将传统的豪斯多夫距离、余弦距离与时间距离相融合,提出一种新的船舶轨迹相似性度量方法。为验证该方法的性能,以宁波-舟山港螺头水道的部分船舶交通数据为例开展对比实验。结果表明,文中提出的新方法在繁忙水道中,相较于豪斯多夫距离、DTW距离、最近邻距离、LCSS距离有着更好的船舶轨迹分类性能,获得的轨迹聚类结果更能反映船舶交通流空间分布特性,可为水上交通规划、交通组织以及航线优化提供参考依据。In recent years,data mining technology has been widely used in the research of waterborne ship traffic,and a large amount of ship trajectory data obtained based on AIS contains ship behavioral features,how to measure the similarity between ship trajectories through these behavioral features in order to extract ship trajectories with similar movement patterns is the focus of the current research and the hot issue.In this paper,a new ship trajectory similarity metric is proposed by integrating the traditional Hausdorff distance,cosine distance and time distance from the characteristics of ship traffic flow.In order to verify the performance of the method,a comparative experiment is carried out with some ship traffic data of the Luotou waterway in Ningbo-Zhoushan harbor as an example.The results show that the new method proposed in this paper has better ship trajectory classification performance than Hausdorff distance,DTW distance,nearest-neighbor distance and LCSS distance in busy waterways,and the obtained trajectory clustering results better reflect the spatial distribution characteristics of ship traffic flow,which can be used as a reference basis for water traffic planning,traffic organization and route optimization.

关 键 词：相似性度量 AIS数据 交通流特征 船舶轨迹聚类 

分 类 号：U675.7[交通运输工程—船舶及航道工程]