Research on Enhancing Road Safety Through Video Analytics & Connected Vehicles

Principal Investigator(s):

Rusheng Zhang, Assistant Research Scientist – University of Michigan Transportation Research Institute (UMTRI)
Henry Liu, Director – Center for Connected and Automated Transportation (CCAT)
Director – University of Michigan Transportation Research Institute (UMTRI)
Director – Mcity
Professor of Civil and Environmental Engineering – The University of Michigan

Project Abstract:
This project develops and demonstrates an end-to-end prototype that integrates roadside video analytics with connected vehicle (CV) technology to enable real-time safety warnings for drivers. Building upon the MSight roadside perception platform, the system detects, tracks, and predicts vehicle and vulnerable road users (VRU) trajectories using infrastructure-mounted cameras, then transmits safety-critical information to vehicles via C-V2X communication. A vehicle-side onboard application processes received messages and delivered timely, intuitive warnings to drivers. Rather than focusing on productization alone, the work uses a prototype-and-field-test approach at Mcity to quantify current technical performance, identify key technology gaps and integration barriers (e.g., detection reliability for VRUs, end-to-end latency, communication constraints, and driver warning usability), and translate findings into prioritized, actionable recommendations. The outcome is a practical assessment of what today’s video analytics + V2X stack can and cannot deliver, and a roadmap of high-impact next steps for DOTs to advance toward deployable, scalable crash-prevention applications.

Institution(s): University of Michigan Transportation Research Institute

Award Year: 2025

Research Focus: Safety

Project Form(s):