ISSN: 1673-825X    Imprint: Chongqing University of Posts and Telecommunications Journal
Volume listings
  • Current volume
  •  
  • Search by keywords:
  • Search by 
  • Keywords
  • To
  • Journal archive
Related Links
考虑道路形状约束的车辆轨迹聚类方法
Vehicle trajectory clustering method considering road shape
DOI:10.3979/j.issn.1673-825X.2019.04.017
Received:March 27, 2018  Revised:January 30, 2019
中文关键词:海量车辆轨迹数据  车辆轨迹  聚类分析
英文关键词:massive data in internet of vehicles  vehicle trajectory  clustering analysis
基金项目:国家自然科学基金(61601126,61571129,U1405251);福建省自然科学基金(2016J01299)
Author NameAffiliationE-mail
FENG Xinxin School of Physics and Information Engineering, Fuzhou University, Fuzhou 350116, P. R. China fxx1116@fzu.edu.cn 
LIU Zefeng School of Physics and Information Engineering, Fuzhou University, Fuzhou 350116, P. R. China njonbbh@163.com 
XIE Zhipeng School of Physics and Information Engineering, Fuzhou University, Fuzhou 350116, P. R. China 263944006@qq.com 
ZHENG Haifeng School of Physics and Information Engineering, Fuzhou University, Fuzhou 350116, P. R. China zhenghf@fzu.edu.cn 
Hits: 25
Download times: 12
中文摘要:
      随着车联网技术的不断发展,产生了海量车辆轨迹数据。这些车辆轨迹数据可以通过聚类分析方法挖掘出车辆行驶的潜在规律,从而实现指导车辆出行的目的。提出一种基于密度的车辆轨迹聚类方法,对基于道路形状关键点位置选取的车辆轨迹信息进行重构,并考虑车辆在路网中移动的空间约束,分析聚类结果得到城市道路的交通状况,以此指导车辆出行以避免或减轻车辆拥堵。基于福州市真实的车辆数据对提出的车辆轨迹聚类算法进行验证,并对最后的聚类结果进行了详细的分析。实验结果表明,针对车辆轨迹聚类并结合道路网络的方法能够更加真实反映车辆的行为特征。
英文摘要:
      With the continuous development of technologies about IoV (Internet of Vehicles), huge amounts of vehicle trajectory data have been generated. These trajectory data can be used to explore the potential rules of vehicle driving through the method of track data clustering, so as to guide the drivers and pedestrians. In this paper, a density based vehicle trajectory clustering method is proposed, in which the vehicle trajectories are reconstructed based on the key points of the road, and the road shapes are taken into consideration. Afterwards, we analyze the urban traffic via the results of clustering to guide the drivers for avoiding the traffic jams. Moreover, we conduct thorough simulation to verify the performance of the proposed vehicle trajectory clustering algorithm based on the vehicle data of Fuzhou city and analyze the final result in detail. The experimental results show that our algorithm can show the characteristics of urban traffic clearly and reflect the behaviors of vehicles accurately.
HTML    View PDF   View/Add Comment  Download reader