一种面向多场景的轨迹相似度计算框架  

A Trajectory Similarity Calculation Framework for Multiple Scenarios

作　　者：赵卓峰[1,2] 鲍越 李舸飞 ZHAO Zhuofeng;BAO Yue;LI Gefei(School of Information,North China University of Technolgy,Beijing 100144,China;Beijing Key Laboratory on Integration and Analysis of Large-Scale Stream Data,Beijing 100144,China)

机构地区：[1]北方工业大学信息学院,北京100144 [2]大规模流数据集成与分析技术北京市重点实验室,北京100144

出　　处：《北方工业大学学报》2024年第1期91-99,共9页Journal of North China University of Technology

基　　金：北京市自然科学基金项目(4202021)。

摘　　要：轨迹相似度计算作为轨迹数据分析中的一个研究热点,主要用来衡量两个移动对象轨迹相似程度。已有的轨迹相似度计算工作主要根据不同类型的轨迹数据和应用场景提出了大量轨迹相似性度量方法,但随着轨迹数据采集手段的逐渐丰富和应用需求的发展,在轨迹相似性计算方面却缺少一个统一的框架以支持多样化轨迹数据下的相似轨迹分析。为此,本文在综合分析已有相似轨迹度量方法共性特征基础上,提出了一个兼容多种轨迹相似度度量方法的集成计算框架(Multiple Scenarios Similarity Trajectory Framework,MSSTJ),并通过一种统一的轨迹数据分区索引方法来优化不同相似轨迹度量方法的计算效率,从而使得用户可以通过参数化配置的简单方式来实现不同场景下多样化的轨迹相似度高效计算的需求。通过基于两类不同类型轨迹数据集的相关实验表明,MSSTJ可以帮助用户快速实现不同的轨迹相似度计算算法,并且实现的算法具有较好的性能表现。Trajectory similarity calculation,as a research hotspot in trajectory data analysis,is mainly used to measure the degree of similarity between the trajectories of two or more moving objects.The existing trajectory similarity calculation work mainly proposes a large number of trajectory similarity measurement methods based on different types of trajectory data and application scenarios.However,with the enrichment of trajectory data collection methods and the development of application requirements,there is a lack of a unified way to supporttrajectory similarity calculation for multiple scenarios.Therefore,based on a comprehensive analysis of the characteristics of existing similar trajectory measurement methods,this article proposes an integrated calculation framework,named as Multiple Scenarios Similarity Trajectory Framework(MSSTJ)that is compatible with multiple trajectory similarity measurement methods,and optimizes the computational efficiency of different similar trajectory measurement methods through a unified trajectory data partitioning index method.This enables users to achieve efficient calculation of diverse trajectory similarity in different scenarios through a simple parameterized configuration way.Experiments based on two different types of trajectory datasets,show that MSSTJ can help users quickly implement different trajectory similarity calculationalgorithms and applications with good performance.

关 键 词：轨迹相似度计算 多场景 集成计算框架 分区索引优化 轨迹相似度算法 

分 类 号：TP393.1[自动化与计算机技术—计算机应用技术]