Reasoning with Uncertain and Incomplete Information in Aerospace Applications (Symposium Paper)

摘要

In many real-life application areas such as aerospace, decisions have to be taken based on imperfect knowledge. If decision makers are to be supported by computer systems, it is desirable that this type of knowledge can be represented. In the past years, new methods have been developed to represent various kinds of imperfectness, such as incompleteness, inexactness or uncertainty. The author investigated the applicability of the Dempster-Shafer theory. He modeled parts of the initiation and identification problems in multi-radar tracking. The application is described in the paper. It is shown that the Dempster-Shafer theory promises improvements over the Bayesian approach, but also that the latter currently is a more advanced theory than the former.