Data fusion methodology for link travel time estimation for advanced traveler information system

Real-time traffic data play a significant role for the provision of real-time traveler information and the development of other ITS (Intelligent Transport Systems) system. For instance, the steady real-time data can be used for incident detection system and dynamic route guidance. Many public transportation agencies or private information service providers are currently or will be providing the road traffic information. The basic form of traffic information would be traffic congestion status, incident, construction area, and the optimum path based on current traffic status. The systems deployed are now evolving with scanty traffic sources that are available. In spite of such poor infrastructure situation, we are trying to derive the link travel times by using whatever information sources available. Currently, the available sources are fixed traffic detectors, CCTVs, and probe vehicles that are running for dispatching and/or for solely collecting the link travel times. The purpose of this paper is to apply the framework of fuzzy operator logic to present methodological aspects of data conversion of each traffic sources and data fusion design by using those traffic data sources, and to implement the fusion mechanism in a typical traffic information center.