Automating Planning Against TSTs via a Novel Combination of Ontologies and Rules

Effectively targeting and engaging time sensitive targets (TST) is an important part of modern warfare. However, doing so remains challenging because of the various considerations that go into deciding whether a given popup is a TST and whether to pursue it. Systems for supporting such mission-critical decision making are crucial in helping to decrease the time for decision making, and to increase the reliability and robustness of the decision making process. We propose an innovative system called Sentinel for helping support the human decision making process in planning against TSTs. Sentinel combines the technologies of ontologies and rule bases in a novel way to address the problems of representation of decision analysis and automation of decision support. In addition, it provides visibility into the steps of the decision making process and takes into account feedback received from the human decision maker. Careful consideration has been given in Sentinel to issues critical to user adoption, namely its usability and its maintenance over time. Phase I prototype development will provide a solid foundation for complete implementation in Phase II.