Detecting risk earlier with connected-vehicle data, computer vision, LiDAR, and emerging mobility sources.
Transportation Researcher · Engineer · Instructor
Saving lives through intelligent transportation research.
I develop smart sensing, transportation digital twins, and innovative analytical systems to identify risk earlier, guide proactive interventions, and make mobility safer for everyone.
Current basePh.D. Candidate in Transportation Engineering · University of Washington . Smart Transportation Application and Research (STAR) Lab
Research lensSafety · Intelligence · Implementation
Why this work matters
Saving lives through transportation research.
I am a transportation researcher at the University of Washington working at the intersection of traffic safety, artificial intelligence, computer vision, simulation, and human-centered mobility.
My work is driven by a simple responsibility: move safety analysis upstream so risks can be understood before they become tragedies. I combine digital twins, AI-enabled smart sensing, and mobility data to support earlier intervention and more defensible transportation policy.
Two decades of national evidence
U.S. road fatalities, 2004–2023
39,345estimated U.S. traffic deaths in 2024NHTSA preliminary estimate ↗1.19 millionroad traffic deaths worldwide each yearWHO Global Status Report on Road Safety 2023 ↗
Chart source: Fatality Analysis Reporting System (FARS) ↗
Connecting sensing, trajectories, simulation, and surrogate safety measures for transparent what-if analysis.
Translating NCHRP Research Report 1152 into practical, data-driven safety workflows.
Building cost-effective tools for traffic, surface, weather, and safety monitoring where data are limited.
Understanding pedestrian and transit crash outcomes through statistical and behavioral analysis.
Recognition
Research and service recognized by the transportation community.
May 2026
Washington State ITE Graduate Student Scholarship
Statewide recognition supporting graduate achievement and continued leadership in transportation engineering.
October 2025
Best Student Poster Award
Recognized for communicating transportation-safety research clearly and effectively to a regional research and practice audience.
Research directions
A connected agenda for safer mobility.
Four complementary research directions connect emerging data and intelligent systems to transportation decisions.
View the complete research portfolio ↗
Proactive traffic safety analytics
Moving safety analysis upstream by connecting trajectories, surrogate safety measures, crash patterns, and machine learning.
Open research story ↗
AI-enabled infrastructure intelligence
Extracting safety-critical roadway and road-user information from imagery, LiDAR, cameras, and multimodal sensing.
Open research story ↗
Transportation safety digital twins
Integrating sensing, trajectories, microscopic simulation, safety metrics, and natural-language interfaces for transparent what-if analysis.
Open research story ↗2026
From Camera-Based Sensing to Reasoning: A Comprehensive Review Toward Proactive Vulnerable Road User SafetyIEEE Transactions on Intelligent Transportation Systems, published online July 20, 2026, 1–17
↗2026Can surrogate safety measures explain crash patterns at signalized intersections? Evidence from network-wide connected-vehicle data using negative binomial, random forest, and diffusion GNN modelsJournal of Safety Research, 97, 553–564
↗2026Leveraging image semantic segmentation for road infrastructure localization in traffic safety applicationsTransportation Research Record, 2680(4), 210–225
↗2026NCHRP Research Report 1152: Leveraging AI and big data to enhance safety analysisTR News, Issue 357, Spring 2026, Transportation Research Board
↗Research in motion
Projects and working demonstrations.
These projects turn research questions into systems, analytical workflows, and tools that can be tested and discussed.
Explore every project ↗University of Washington
↗AI-Enhanced Traffic Safety Digital Twin
A research platform connecting road-user sensing, SUMO simulation, safety analytics, comparative visualization, and a natural-language interface.
- Digital twin
- SUMO
- LLM
- Safety analytics
University of Washington · Tribal Transportation Safety Collaboration
↗Tribal and Rural Traffic-Safety Digital Twin
A cost-effective sensing, simulation, and communication system designed to move rural and tribal safety management from observation toward proactive intervention.
- Tribal transportation
- Rural safety
- Digital twin
- Smart sensing
- Local LLM
NCHRP 17-100 · Research Report 1152
↗Leveraging AI and Big Data to Enhance Safety Analysis
Machine-learning applications across connected vehicles, street-view imagery, LiDAR, speed prediction, road-user behavior, and surrogate safety measures.
- Big data
- Computer vision
- Connected vehicles
- Safety analysis
A journey from practice to research
Engineering questions shaped by the real world.
My path in transportation began in professional practice, where I worked on freight, multimodal planning, traffic-impact, road-safety, and master-planning projects. Those experiences taught me that a useful method must work beyond the page: it must respond to the constraints, uncertainty, and public responsibilities of real decisions.
Field notes
Conferences, collaborations, and moments along the way.
Photo collections are organized by event so the people and exchanges behind the research have their own place.
Browse the photo journal ↗
Continue the conversation
Have a research question, dataset, or idea worth exploring?
I welcome conversations about transportation safety, intelligent mobility systems, research demonstrations, teaching, and interdisciplinary collaboration.
mehnasri@uw.edu