COAXIS - Course of Action Extra-Intelligence System Navy STTR FY2014.A Sol No.: Navy STTR FY2014.A Topic No.: N14A-T024 Topic Title: COAXIS - Course of Action Extra-Intelligence System Proposal No.: N14A-024-0033 Firm: Harmonia Holdings Group 2020 Kraft Drive, Suite 1000 Blacksburg, Virginia 24060-6491 Contact: Marc Abrams Phone: (540) 951-5901 Web Site: www.harmonia.com Abstract: COAXIS (Course of Action Extra-Intelligence System) recommends both information relevant to an operational order and a course of action (COA), where COA is defined as a decision of what action to take given a specific situation. COAXIS performs semantic analysis of natural language of an operational order to bridge the world of human-oriented natural language communication of orders to machine analysis and prediction of events on the battlefield to recommend COAs. We solve the problem in a top-down, which allows us to address the four challenge areas in the topic. We start with a reasoning engine to make COA recommendations using information about assets, environment (adversary, neutrals, METOC, network, spectrum,...), geography, time, and mission in terms of capability requirements, tasks, goals (measures of performance and effectiveness). To make recommendations, we then address what queries will route data from the right sources. That in turn raises the question of what data is needed, which we address by mapping orders and current context to formulating the queries. Finally the orders themselves come from natural language analysis of human created messages. Our solution builds on our existing work in fusion, layered data, and battle management aides. Benefits: COAXIS allows faster decisions, based on consideration of more variables when making recommendations than a human decision maker could consider in their mind. COAXIS is not meant to be a silver bullet that replaces human decision; instead it is a tool to augment an experienced decision maker and bring to the surface possible situations with disadvantages to jog the human decision maker's thinking. COAXIS will learn through training, so that it does not require an expensive knowledge elaboration process to create a rule base. The ability in COAXIS to do projection of the current timeline provides an ability to adapt COAs to possible future events, making the solution more dynamic. Return