Monitoring and control of extreme congestion events in road traffic networks

This thesis addresses the problem of traffic management during extreme traffic congestion events. In particular, it describes the development of a new traffic sensing technology called TrafficTurk that is capable of being deployed through multiple temporary sensors quickly over an entire road network. The development of the technology is motivated by the enormous cost to society that is associated with extreme traffic congestion caused by pre-planned events such as sporting and cultural events, extreme weather and natural disasters. The eventual goal of the technology is to provide a comprehensive system for traffic management of extreme congestion events through real-time data collection and processing. Therefore, detailed explanations of data processing algorithms that are required to enable further development of the system are also provided. Specifically, a method to estimate traffic controller strategies at intersections in a road network using inverse optimal control and a method to normalize traffic data collected using TrafficTurk are explained in detail. Additionally, practical implementation knowledge gained thus far through extensive field testing of the system is also discussed