Real Time Risk Prediction at Signalized Intersection Using Graph Neural Network

Intersection related traffic crash and fatalities are one of the major concerns for road safety. In this project the team aims to understand the major cause of conflicts at an intersection by studying the intricate interplay between all the roadway agents. The project team proposes to use the current traffic camera systems to automatically process traffic video data. As manual annotation of video datasets is a very labor-intensive and costly process, A system that can process these traffic datasets automatically would strongly enhance the effectiveness of the analysis and enable new research questions to be addressed. Therefore, the project team proposes to use computer vision algorithm to process the videos. Also, the team proposes to use advanced machine learning methods including graph neural network (GNN) to model the interaction of all the roadway agents at any given instance, and their role in road safety, both individually and as a composite system. As a result, the proposed model aims to develop a near real time risk score for a traffic scene.

Language

• English

Project

• Status: Active
• Funding: $320000
• Contract Numbers:

  69A3551747115

• Sponsor Organizations:

  Office of the Assistant Secretary for Research and Technology

  University Transportation Centers Program
  Department of Transportation
  Washington, DC  United States  20590
• Managing Organizations:

  Safety through Disruption University Transportation Center (Safe-D)

  Virginia Tech Transportation Institute
  Blacksburg, VA  United States  24060
• Project Managers:

  Glenn, Eric

• Performing Organizations:

  Virginia Tech Transportation Institute

  3500 Transportation Research Plaza
  Blacksburg, Virginia  United States  24061
• Principal Investigators:

  Sarkar, Abhijit

• Start Date: 20220501
• Expected Completion Date: 20230630
• Actual Completion Date: 0
• USDOT Program: University Transportation Centers

Subject/Index Terms

• TRT Terms: Crash risk forecasting; Highway safety; Image processing; Real time data processing; Signalized intersections; Video cameras
• Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management; Safety and Human Factors;

Filing Info

• Accession Number: 01842000
• Record Type: Research project
• Source Agency: Safety through Disruption University Transportation Center (Safe-D)
• Contract Numbers: 69A3551747115
• Files: UTC, RIP
• Created Date: Apr 7 2022 4:25PM