DISCOVERY MACHINE INC — Department of Defense SBIR Phase II: AF221-DCSO1

Description

Discovery Machine®, Inc. (DMI) proposes to provide voice-enabled cognitive agents for two Remotely Piloted Aircraft (RPA) training needs as a part of the Undergraduate RPA Training Next (URT-N) program and in support of RPA mission training at the 25th Attack Group. Here we propose the Sensor Coach for Optimizing RPA Pilots Intelligent Operational Needs (SCORPION™) as a virtual instructor to carry out these training needs. DMI’s solution leverages our commercial software including DMI’s Knowledge Service Engine (KSE), the Population Composer, and the Rapidly Encoded Simulation Immersive Training Environment (RESITE®) Suite. DMI shall also leverage a set of extant cognitive models created for Airforce Education Training Command’s (AETCs) Pilot Training Next (PTN) program used in the Virtual Instructor Pilot Exercise Referee (VIPER®), as well as, cognitive agents developed for the U.S. Marine Corp in Patterns of Life (PoL) agents deployed in RESITE®. If time permits, the addition of models created for the Joint Theater Air Ground Simulation System (JTAGSS) program shall also be used to create air traffic, chatter, and realistic air tasking. DMI shall apply our Hybrid AI techniques to build intelligent agents to support MQ-9 tactical and terminal area training. Discovery Machine® shall build intelligent agents to increase the realism and immersion of MQ-9 pilots training in both tactical and terminal areas. Additionally, Discovery Machine® will produce as needed synthetic external entities with which MQ-9 pilots must collaborate with to accomplish their objectives. To ensure the development of both doctrinal and subject matter expert validated behavior, Discovery Machine® will also provide an intelligent tutoring agent to observe and coach during training exercises. A combination of all these tasks will garner the measurable goal of reducing the time and skill requirements of white force sim operators.