Vehicle trajectory Search

Project overview

Torch is a new trajectory search engine for querying road network trajectory data. Torch is able to efficiently process two typical types of queries (similarity search and Boolean search), and support a wide variety of trajectory similarity functions. Beyond that, in Torch a new similarity function LORS is proposed to measure the similarity in a more effective and efficient manner. Torch mainly works as below. First, each raw vehicle trajectory is transformed to a set of road segments (edges) and a set of crossings (vertices) on the road network. Then a lightweight edge & vertex index called LEVI is built. Given a query, a filtering framework over LEVI is used to dynamically prune the trajectory search space based on the similarity measure imposed. Finally, the result set (ranked or Boolean) is returned.We devise an evaluation method, and extensive experiments on real trajectory datasets verify the effectiveness and efficiency of Torch, and the LORS driven by LEVI.

  • Pre-processing & Index
    • Map matching raw data to road network
    • Robust to sampling rate
  • Query
    • Path query
    • Strict path query
    • Range query
    • Similarity search
  • Evaluation
    • Simulation trajectories using different sampling rate

Prototype

Datasets

Supported publications

  • Sheng Wang, Zhifeng Bao, J. Shane Culpepper, Timos Sellis, Xiaolin Qin. Fast Large-Scale Trajectory Clustering. VLDB 2020, Tokyo, Japan. (To appear)
  • Sheng Wang, Zhifeng Bao, J. Shane Culpepper, Zizhe Xie, Qizhi Liu, Xiaolin Qin. Torch: A Search Engine for Trajectory Data. ACM SIGIR 2018, Ann Arbor Michigan, U.S.A.
  • Sheng Wang, Yunzhuang Shen, Zhifeng Bao, Xiaolin Qin. Intelligent Traffic Analytics: From Monitoring to Controlling. ACM WSDM Demo 2019, Melbourne, Australia.