The Washington State Department of Transportation (WSDOT) operates a central traffic management system (TMS) for both day-to-day surveillance and traveller information. Past efforts developed the ability to create real-time traffic speed information by using virtual sensors that are based on transit vehicle tracking data. In order for this new information source to be merged into the TMS, a number of questions, such as probe density in time and space, needed to be resolved. This report presents the solution developed at the University of Washington (UW). This solution provides real-time congestion information from Seattle area freeways and arterials--I-5, I-90, SR 520 and SR 99--to the WSDOT TMS using the Intelligent Transportation System (ITS) Backbone. This project harvests existing automatic vehicle location (AVL) data from within King County Metro Transit and transports the raw data to the UW, where a series of operations converts the data into roadway speed information. This roadway speed information is colour-coded on the basis of specific, localized conditions for the arterials and freeways to reflect traffic congestion. The resulting traffic data product is then provided to WSDOT as a data source for virtual sensors located in roadways where currently there are no inductance loops. In addition to creating the infrastructure for an AVL-equipped fleet to serve as probe vehicles, this project created several user interfaces for traveller information. One is "StoreView," a Java application that displays the spatial and temporal average speeds of transit vehicles as colour-coded bubbles on a map of the area's major arterials and freeways. A second type of traveller information, analogous to TrafficTV and WSDOT's pictographic traffic maps, is also available. This report documents both the technical issues addressed in creating a virtual sensor data stream from probe vehicle data and the creation of a set of real-time traveller information applications.
Samenvatting