Network Topology Aware Moving Object Trajectory Clustering
This paper illustrates a three step clustering method for moving object trajectories to discover common sub-paths. The proposed algorithm (TOPOSCAN) is inspired from the well-known partition-and-group framework and density based algorithms. It takes advantage of topological relation of a predefined network, and reduces the dimension of trajectory databases and algorithm complexity. TOPOSCAN consists of three steps: network topology construction, trajectory mapping, and path clustering. For network topology construction, we define the network as a directed graph, associated with a network attribute table to describe link relationships. For the trajectory mapping, the authors adopt a map matching algorithm to assign trajectory points to network links. For the path clustering, the authors develop a framework with extraction, extension and exclusion steps to select the common sub-paths. TOPOSCAN is implemented using simulated trajectory data. The experimental results demonstrate that TOPOSCAN correctly and efficiently discovers common sub-paths from vehicle trajectory data. The recognized patterns are useful for transport management and operation design.
-
Supplemental Notes:
- This paper was sponsored by TRB committee ABJ70 Standing Committee on Artificial Intelligence and Advanced Computing Applications.
-
Corporate Authors:
500 Fifth Street, NW
Washington, DC United States 20001 -
Authors:
- Hong, Zihan
- Xu, Shuang
- Mahmassani, Hani S
-
Conference:
- Transportation Research Board 95th Annual Meeting
- Location: Washington DC, United States
- Date: 2016-1-10 to 2016-1-14
- Date: 2016
Language
- English
Media Info
- Media Type: Digital/other
- Features: Figures; References; Tables;
- Pagination: 16p
- Monograph Title: TRB 95th Annual Meeting Compendium of Papers
Subject/Index Terms
- TRT Terms: Algorithms; Cluster analysis; Data mining; Topology; Trajectory
- Subject Areas: Data and Information Technology; Highways; I72: Traffic and Transport Planning;
Filing Info
- Accession Number: 01590392
- Record Type: Publication
- Report/Paper Numbers: 16-6968
- Files: TRIS, TRB, ATRI
- Created Date: Feb 17 2016 11:59AM