Real-time Pedestrian Safety and Risk Exposure using Real-time Vehicle Activity and Fleet Composition

Pedestrians face significant a risk crossing roadways as they interact with vehicle traffic (more than 7,000 pedestrians were killed in traffic crashes in 2023).  New tools that assess risk exposure can improve the safety of routes delivered by navigation apps.  In 2025, the research team generated a complete-paths pedestrian network for Downtown Atlanta and inspected the condition of all sidewalk surfaces.  In 2024, Georgia Tech researchers also began collecting very consistent vehicle images using portable high-resolution video cameras positioned on Interstate overpasses (more than one-million vehicle images) for the State Road and Tollway Authority.  A large subset of vehicle images were coded by vehicle make-and-model and used in a prior research project to develop machine-vision artificial intelligence (AI) models to generate fleet composition profiles, for use in energy and safety research.  In this new project, the researchers will integrate traffic operations data and assess pedestrian exposure to high traffic volumes, high vehicle speeds, and turn movements that cross pedestrian paths.  The team will enhance SidewalkSim and G-MAP (www.its.dot.gov/research-areas/ITS4US/), models that find the “shortest path” (i.e., lowest impedance path) for pedestrian trips between any origin-destination, by integrating traffic exposure into routing impedance factors.  This will allow the apps to route pedestrians around high-risk-exposure crossings.  The team will also map and integrate pedestrian safety countermeasures (bollards, barriers, pedestrian fences, extended crossing times, leading pedestrian intervals, no-crossing zones, no-right-turn-on-red, etc.) in the study area, so that these countermeasures can also be used in impedance-based pedestrian routing along safer paths.  The project culminates by integrating traffic conditions, fleet composition, and risk exposure into SidewalkSim and G-MAP pedestrian routing app and demonstrating the system in downtown Atlanta.  Another finding from past research was that the resulting machine vision models are so fast, they can run in real-time.  In this new project, the research team will further refine the AI models so that they can be used in edge-computing, processing vehicle fleet composition in the field, without transmitting video data to a data center.  In this project, the team aims to design and package an efficient portable computing system with a high-end graphics card that can operate under year-round Atlanta outdoor temperature and humidity conditions, balancing system performance with power-draw and heating/cooling requirements.  This equipment research (downsizing, enclosure design, heat dissipation, power consideration, etc.) and machine vision model implementation may lead to patentable inventions or licensable software.  If successful equipment deployments are afforded patent protection, the team will work with Georgia Tech’s commercialization office (commercialization.gatech.edu) to develop license agreements for the manufacture of equipment and deployment of portable edge-computing systems and/or will create a GT Create-X business startup.  If the USPTO rejects the patent claims, the team will release equipment specifications, software code, and technology transfer reports under open-source licensing that will allow state DOTs and their consultants to implement the systems. 

Language

• English

Project

• Status: Active
• Funding: $383,000.00
• Contract Numbers:

  69A3552348329

• 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:

  Center for Efficient Mobility

  1111 Rellis Parkway
  Bryan, Texas  United States  77807
• Project Managers:

  Ocon, Monica

• Performing Organizations:

  Georgia Institute of Technology, Atlanta

  790 Atlantic Drive
  Atlanta, GA  United States  30332-0355
• Principal Investigators:

  Guensler, Randall

• Start Date: 20260201
• Expected Completion Date: 20270514
• Actual Completion Date: 0
• USDOT Program: University Transportation Centers
• Source Data: 03-09-GT

Subject/Index Terms

• TRT Terms: Machine learning; Mobile applications; Navigational aids; Pedestrian safety; Risk assessment; Routes and routing; Traffic volume; Wayfinding
• Geographic Terms: Atlanta (Georgia)
• Subject Areas: Data and Information Technology; Operations and Traffic Management; Pedestrians and Bicyclists; Planning and Forecasting; Safety and Human Factors;

Filing Info

• Accession Number: 01979472
• Record Type: Research project
• Source Agency: Center for Advancing Research in Transportation Emissions, Energy, and Health (CARTEEH)
• Contract Numbers: 69A3552348329
• Files: UTC, RIP
• Created Date: Feb 15 2026 4:27PM