High-Resolution Traffic Sensing with Autonomous Vehicles

The project explores how autonomous-vehicle fleets could function as mobile sensors to gather traffic data across road networks. We developed methods for extracting traffic insights — including travel speeds, vehicle density, and flow counts — by analyzing object-detection and tracking data from self-driving vehicles.

The team validated the approach using Uber's autonomous-vehicle fleet operating on Pittsburgh surface streets. The work addresses a critical question for transportation planning: how would transportation stakeholders design, plan, and operate infrastructure in the era of automated vehicles?

The research demonstrates that, when sufficiently dense in space and time, AV-derived observations can reveal meaningful patterns about network performance — though this must be weighed against the operational costs of managing large-scale data collection and security.