STEER AV - Safety Tuned Emulation of Emerging Responses for Autonomous Vehicles

Autonomous vehicles (AVs) produced by different manufacturers often display distinct driving styles because each system uses its own proprietary decision rules. These variations can affect safety and traffic flow during the long transition period when automated and human driven vehicles operate together. This project will study real world AV trajectory data to assess how AVs follow other vehicles, how they balance safety and efficiency, and which factors shape their decision making. The research team will use inverse reinforcement learning and interpretable generative methods to infer the policies that guide AV actions and to create models that reproduce these behaviors. After the initial models are created, the project will incorporate additional constraints that guide the system toward safer behavior while preserving mobility. Simulation experiments will examine how these modified policies perform under a range of conditions and will evaluate possible trade offs between safety and efficiency. The resulting framework will support efforts to standardize and improve AV driving policies, help researchers understand AV decision patterns, and assist agencies and manufacturers as they prepare for increasing levels of automated travel.

• Record URL:
  • https://www.umass.edu/neutc/projects/steer-av-safety-tuned-emulation-emerging-responses-autonomous-vehicles

Language

• English

Project

• Status: Active
• Funding: $199,000.00
• Contract Numbers:

  69A3552348301

• 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:

  University of Massachusetts, Amherst

  Department of Civil and Environmental Engineering
  130 Natural Resources Road
  Amherst, MA  United States  01003
• Performing Organizations:

  University of Connecticut, Storrs

  Connecticut Transportation Institute
  270 Middle Turnpike, Unit 5202
  Storrs, CT  United States  06269-5202
• Principal Investigators:

  Filipovska, Monika

• Start Date: 20260101
• Expected Completion Date: 20261231
• Actual Completion Date: 0
• USDOT Program: University Transportation Centers Program
• Subprogram: University Transportation Centers

Subject/Index Terms

• TRT Terms: Autonomous vehicles; Car following; Decision making; Machine learning; Standardization; Traffic safety; Vehicle trajectories
• Subject Areas: Highways; Operations and Traffic Management; Planning and Forecasting; Safety and Human Factors; Vehicles and Equipment;

Filing Info

• Accession Number: 01973939
• Record Type: Research project
• Source Agency: New England University Transportation Center
• Contract Numbers: 69A3552348301
• Files: UTC, RIP
• Created Date: Dec 11 2025 1:33PM