Description: SDi is developing a sophisticated, autonomous artificial intelligence model that is designed to act as a modern-day battlefield commander. STRAT_AGENT utilizes a hierarchy of intelligent agents in combination with a rule-driven expert system (military doctrine) that acts autonomously, directing its activity toward achieving objectives in the STRATAGEM wargame environment. The “intelligence” that STRAT_AGENT utilizes is both distilled from standard military practice and learned through a ’new rule’ creation process driven by cognitive agents.
Primary Objective: STRAT_AGENT is designed to defeat a competing BLUFOR AI or Human in the STRATAGEM wargame.
Innovation: STRAT_AGENT is designed to leverage advances in the use of game AI systems in order to develop an AI system capable of defeating a human expert. Additionally, STRAT_AGENT has the ability to not only win, but also to provide an interpretation of how it won and what logic it used to do so.
SDi has designed a real-time simulation based on a cognitive architecture of intelligent agents (modeling manned military systems) combined with a swarm intelligence architecture (modeling unmanned systems).
SDi is developing an innovative way to enhance the capability of both OPFOR and BLUFOR wargame participants by providing a platform that will complement their own operational abilities.