DEMPSTER-SHAFER DATA FUSION AT THE TRAFFIC MANAGEMENT CENTER
Lawrence A. Klein · 2000
This paper is concerned with the application of Dempster- Shafer inference at a traffic management center (TMC) to support incident detection and the identification of other events of concern to traffic managers. Dempster- Shafer inference, a statistical- based data fusion classification algorithm, is used when the sensors or other data sources contributing information cannot associate a 100 percent probability of certainty to their output decisions. The algorithm captures and combines whatever certainty exists in the object discrimination or event classification capability of the sensors and other sources of information. Knowledge from multiple sources about the events is combined using Dempster's rule to find the intersection or conjunction of the events and the associated probability. The application of the algorithm to incident detection and verification is illustrated with an example consisting of three possible events, where data are supplied from three different types of sources. The available information is combined using Dempster's rule and the most probable event is identified