Multi-Source Data Fusion for Urban Traffic State Estimation: A Case Study of New York City

Details

• Creators:
  Gao, Jingqin ; Ozbay, Kaan ; Kurkcu, Abdullah
• Corporate Creators:
  Connected Cities for Smart Mobility toward Accessible and Resilient Transportation Center (C2SMART)
• Subject/TRT Terms:
  • Traffic flow
  • Data fusion
  • Traffic estimation
  • Traffic flow theory Traffic state estimation
  • Virtual sensors
  • Evidence theory
• Resource Type:
  Manuscript
• Geographical Coverage:
  United States;New York (New York)
• Corporate Publisher:
  Connected Cities for Smart Mobility toward Accessible and Resilient Transportation Center (C2SMART)
• Abstract:
  Data fusion techniques are often used to enhance traffic state estimations. The objective of this paper is to evaluate and validate the applicability of three different data-driven fusion techniques over a non-trivial urban transportation network. Simple weighted, machine learning and evidence theory based approaches are applied to generate estimate for traffic states. Multiple data sources including information collected from electronic toll collection tag readers, Global Positioning System-equipped probe vehicles, and crowdsourcing map applications are utilized in the study. A case study is provided to illustrate an application of the fusion techniques with data extracted in 2017 for two weeks. Ground truth information collected from real-time camera feeds are used for validation. The evidence theory based method considering temporal evidence reliability outperforms the other methods in terms of the cross validation of the model accuracy. In addition, this study proves that the information extracted from web-based map services using “Virtual Sensors” can be an excellent supplementary data source for current travel time monitoring systems at no additional cost.
• Format:
  PDF
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/43892/dot_43892_DS1.pdf
• File Type:
  [PDF - 877.50 KB]
• Collection(s):
  University Transportation Centers ; US Transportation Collection
• Main Document Checksum:
  urn:sha-512:c5b0ec05f1cbb4704d07b706a17eba1cfa7f12829ce2b56069f5b39238299a681637f6931503aca8b99181c26570fcea77618777e8426e73c8ef574c82c68802