Monitoring Traffic During Extreme Congestion Events

09:00 AM - 10:00 AM
Daniel B. Work, The University of Illinois, Urbana-Champaign
National Transportation Research Center Seminar
NTRC-2, TRAVL (Room 111)
Email: Debbie Bain

Dr. Work will discuss a new smartphone application , Traffic Turk, that allows for low-cost, large-scale and real-time traffic sensing, estimation and control. Traffic Turk has been tested during natural; disasters (Hurricane Sandy) and a major sport event. He will de¬scribe a new approach to monitor traffic with cell phones, known as TrafficTurk. Inspired by Amazon's Mechanical Turk for crowd--sourcing human intelligence tasks, TrafficTurk enables temporary large-scale traffic sensor deployments to improve data coverage during extreme congestion events.

TrafficTurk emulates the simple but surprisingly effective manual counting devices traffic movement counters used by most municipalities to periodically measure traffic counts at intersections, but with streaming data capabilities and cheap deployment costs.

In addition to improving data coverage, TrafficTurk also enables traffic control strategies to be recov¬ered by solving an inverse optimal control problem on the observed data. Experimental deployments in New York following Hurricane Sandy and a 100--node test in Urbana will be discussed.


We're always happy to get feedback from our users. Please use the Comments form to send us your comments, questions, and observations.