摘要: |
This report documents the second phase of a three-phase project that will create a robust set of virtual sensors for freeways and arterials. The first phase was a proof of principle that examined the statistics of successfully using transit vehicles as traffic probes. The results of the second phase are presented in this report. An optimal filter method is described that estimates acceleration, speed, and position as a function of space and time. The third phase will implement a server to place speed estimates from the transit probe virtual sensors into the Northwest Region's operational Traffic Management System. The overall project will accomplish several goals, it will: (1) create a mass-transit vehicle tracking system that computes smooth estimates of speed and distance traveled for each vehicle in the fleet, (2) create a network of virtual 'speed sensors' on selected road segments by using the 'probe' vehicle speed estimates produced by the tracking system, (3) make the speed measurements available for traffic analysis purposes and data fusion with CCTV camera and inductance loop measurements, (4) create graphical applications for visualization of current and historical traffic conditions, (5) create tools for predicting pointto- point travel times on the basis of smoothed historical speed data, and (6) create a corridor speed estimate and travel time. |