|Presentation Date||October 30, 2012|
|Topic(s)||Traffic Ground Truth Estimation Using Multisensor Consensus Filter|
Estimation of traffic ground truth has traditionally been accomplished either by the use of a trusted reference detector, or by human observation of time-coded video recordings. These approaches are limited by the error characteristics of any single trusted detector, or the temporal and spatial resolution of the recorded video used by the human observer. For conventional traffic management purposes, the output of a trusted detector is usually adequate. However, for unbiased performance assessment and comparative ranking of vehicle detectors, an accurate ground truth estimate is essential. With this objective, Dr. Art MacCarley and I implemented an automated detector testing system which utilizes consensus filtering methods typically employed in the fusion of data from multi-sensor networks to optimally estimate ground truth. The algorithm continuously adjusts a level of confidence in each detector under test, which are used to form a weighted consensus decision for the presence, speed and length of each vehicle. Individual detectors under test are then assessed by comparison with the estimated ground truth record, and by the confidence factors generated by the algorithm. I will describe the algorithm and present testing methodology to determine its ability to accurately estimate ground truth by use of synthetic traffic data for which absolute ground truth is known, and synthetic detector data derived from the ground truth with known injected errors. I will also describe the application of this algorithm in the Caltrans Advanced Traffic Management Systems Detector Testbed.