Flow speed describes general traffic operation conditions on a segment of roadway. It is also used to diagnose special conditions such as congestion and incidents. Accurate speed estimation plays a critical role in a traffic management or traveler information system. Data from loop detectors has been a primary source for traffic information, and single loop detectors are the predominant source in many places. However, single loop detectors do not produce speed output. Several methods have been developed for speed estimation using single loop detector outputs. These methods, however, have their limitations and are often inaccurate under various traffic conditions. Some of the methods are also difficult to implement. This research project seeks to improve on the existing methods and to increase the accuracy of speed estimation. A new methodology, the Unscented Kalman Filter (UKF) method, is developed for this purpose. Datasets collected from three different freeway locations are used for speed estimation and evaluation of the proposed method. The results show that the proposed method generates accurate and stable estimations of speed. The proposed method is superior to existing methods. (Author/publisher)
Abstract