Applying AI to data sources to improve driver-pedestrian interactions at intersections

This project aims to understand how we can link and harness new data sources along with machine-learning based optimization techniques to improve driver-pedestrian interactions at intersections. With a strong emphasis on safe mobility, this study will address this critical issue in a real-life context with data links from and actuation feedback to eight monitored intersections in the Chattanooga Shallowford road corridor between the Lee Highway and Gunbarrel Road intersections. The study reflects strong collaborations between UTK, UNC, ORNL, and the City of Chattanooga, TN. This project will contribute by incorporating safety into optimization of traffic signals, collect and link data from traffic signals (cameras) and analyze behaviors of pedestrians and drivers at intersections, and provides an opportunity for students to engage in multiple aspects of safety analysis including data linking and new AI techniques. The impact of this project can be substantial in terms of enhancing safety and efficiency of intersections. Working closely with Chattanooga and ORNL, the project team will impact the safety as well as performance of traffic signals in a testbed corridor through: 1) data linking, 2) pedestrian detection and 3) optimization.

• Record URL:
  https://www.roadsafety.unc.edu/research/projects/2021r43/
• Supplemental Notes:
  • CSCRS2021R43

Language

• English

Project

• Status: Active
• Funding: $67500
• Contract Numbers:

  69A3551747113

• 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:

  Collaborative Sciences Center for Road Safety

  University of North Carolina, Chapel Hill
  Chapel Hill, NC  United States  27514
• Project Managers:

  Sandt, Laura

• Performing Organizations:

  University of Tennessee, Knoxville

  Knoxville, TN  United States 

  University of North Carolina at Chapel Hill

  UNC-CH New East Building
  Campus Box #3140
  Chapel Hill, North Carolina  United States  27599-3140

  Oak Ridge National Laboratory

  545 Oak Ridge Turnpike
  Pittsburgh, TN  United States  37830
• Principal Investigators:

  Chakraborty, Subhadeep

  Khattak, Asad

  Nordback, Krista

  Sanyal, Jibonananda

• Start Date: 20210501
• Expected Completion Date: 20230831
• Actual Completion Date: 0
• USDOT Program: University Transportation Centers Program

Subject/Index Terms

• TRT Terms: Artificial intelligence; Behavior; Drivers; Intersections; Machine learning; Optimization; Pedestrian detectors; Pedestrian vehicle interface; Pedestrians; Traffic safety; Traffic signal control systems
• Geographic Terms: Chattanooga (Tennessee)
• Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management; Pedestrians and Bicyclists; Safety and Human Factors;

Filing Info

• Accession Number: 01771372
• Record Type: Research project
• Source Agency: Collaborative Sciences Center for Road Safety
• Contract Numbers: 69A3551747113
• Files: UTC, RIP
• Created Date: May 7 2021 10:28AM