Seminar: AI-empowered Digital Twin for Traffic Safety Analysis

Abstract: Traffic safety research faces the paradox of rarity—the most critical crashes occur so infrequently that passive methods relying on historical records cannot accurately estimate traffic risks. This talk first reviews current methodologies and highlights their limitations in handling rare, high-risk events. Building on these insights, we chart a new path that fuses active safety analysis of near-miss events and generative AI. As a preliminary foundation for the generative AI–enabled safety analysis, our recent work provides digital-twin–enabled active safety analysis through three pivotal components: (1) a physics-grounded active safety module that identifies near-miss events across diverse traffic contexts; (2) group-wise interaction modeling that captures multi-agent behaviors in traffic; and (3) a high-fidelity digital twin integrating detailed vehicle dynamics, tire–road interaction, and the above active-safety and interaction models. Together, these elements enable proactive risk prevention, expanding safety analysis beyond observed outcomes to the full spectrum of what could happen, and lay the foundation for Vision Zero.

Bio: Dr. Yang Zhou is an assistant professor in the Zachry Department of Civil and Environmental Engineering at Texas A&M University, and career initiation fellow of Texas A&M Institute of data science. He received his Ph.D. degree and Master degree from the University of Wisconsin-Madison and the University of Illinois at Urbana and Champaign, respectively. Before joining Texas A&M, Yang worked as a postdoctoral research associate supported by the Department of Civil and Environmental Engineering, University of Wisconsin-Madison.  Yang has over ten years of experience in connected automated vehicle control and analysis, traffic flow analysis, AI applications on transportation, and high-fidelity simulation. Yang has PIed multiple federal and local grants such as FHWA-EAR and SS4A. Yang has published more than 70 top-tier transportation journals, including Transportation Research Part B, Transportation Research Part C, and IEEE Transactions on Intelligent Transportation Systems.