Performance comparison of GPS probe-vehicle-based methods in urban traffic state estimation

Global Positioning System (GPS)-equipped probe vehicles have been one of the principal techniques in detection of traffic flow. In order to guide the application of the different methods using data from GPS probe vehicles, this paper compares two of the most popular methods using the GPS data to estimate traffic state- the method of tracking probe vehicles, and the one of fitting the model with the sample points of probe vehicles. In the experiment, data from thousands of GPS-equipped taxies were taken as the probe vehicles, and the estimate accuracy and operation speed of the two different methods were systematically measured. For the accuracy experiment, the ground truth was obtained by replaying the videos which were shot on 24 links in Shanghai downtown. The experiment results illustrate that the tracking-based method usually bears higher estimate accuracy, and slower operation speed, compared to the model-fitting method. © 2011 ASCE.