Use of Crew Intent Inferencing to Augment Control of Multiple Unmanned Aerial Vehicles

"Unmanned aerial vehicles are valuable assets to the Army's scout and attack rotorcraft on the digital battlefield. Intent inferencing is critical for these types of mixed-initiative systems designed to carry out the commander's intent without violatingmission constraints or overloading their airborne human supervisors. The Monterey Technologies, Inc. (MTI) team has directly applicable experience designing, building, and evaluating intelligent systems for medical diagnosis, tutoring, and militarytactical planning. Our approach will use an agile modeling techniques to rapidly design an RPA and Apache Longbow-compatible intent inferencer that accurately and reliably estimates human operator activities and goals based on pilot actions, mission goalsand constraints, and aircraft state, and employs state of the art techniques for resolving uncertainty with minimal operator interruption. Our modeling techniques will allow us to rapidly identify, document, and communicate system requirements andimplementation constraints, focusing the majority of design time addressing the context model and the challenge of reasoning under uncertainty. MTI will leverage recent technological advances in KBS development approaches such as Bayesian networks, casebased reasoning, and hybrid architectures integrating rule and case-based inferencing. The functionality and portability of our final HLA-compliant intent inferencing system design will be demon