How certain is certain? Evaluation of uncertainty in the fusion of information derived from diverse sources

Abstract

Todays asymmetric military operations, often involve multiple partners, including the armed forces of allied nations, relief agencies and other organizations. Background information is culled from many different sources, including humans, sensors, and robots. Timely evaluation of intelligence with background information is necessary for effective operations, but its sheer volume poses a challenge. Formulating reports involving actions, events and status in a standardized formal language such as battle management language (BML) provides a basis for automatic fusion of individual pieces of information into complex, identifiable patterns of behavior. However, since sources of information are seldom completely reliable, the fusion of data is inherently unreliable. Evaluating the degree of reliability of the information will support effective decision making. In this paper we examine issues involved in the evaluation of the uncertainty in the fusion of information from diverse sources and propose a methodology for evaluation of this uncertainty.