Cognitive and Behavioral Modeling Techniques for CGFs: A New Initiative

摘要

A recent NRC report identified that a shortcoming in the area of military simulations is the lack of behavioral realism in computer generated forces (CGFs). Given the military's growing reliance on large-scale simulations as a means to prepare our warfighting teams, this deficiency has far reaching consequences. Therefore, ONR and NAWCTSD are launching a research initiative to investigate cognitive and behavioral modeling (CBM) techniques suitable for injecting into executable models of combatant behavior. In particular, it is necessary to devise robust strategies for simulating higher-order processes in CGFs, including decision making, intent, deception, adaptability, creativity, and problem solving. This initiative will be executed in three phases: Phase 1 - - Develop a taxonomy that describes military simulation applications and modeling methodologies with a common set of discriminating features/attributes to study the technical capabilities and shortcomings of current CGFs and CBMs; Phase 2 -- Convene a multi-disciplinary panel of experts at a workshop to validate the findings and provide feedback (the results will be published as guidelines to advance, implement and apply CBM techniques for CGFs); and Phase 3 -- Based on the initial findings, this initiative will facilitate the application of CBM methodologies currently viable for military simulation, implement hybridization of CBM techniques that provide part of a solution, and invest in CBM advancements where insufficient theory exists.