Defensive Coordinator for Autonomous Countermeasure Systems

Current knowledge and experience in connection with potential conflict scenarios involving Unmanned Air Systems (UAS) is limited. There exists little or no data to inform us how to respond even to situations we can imagine (such as those in which potentially vast numbers of hostile UAVs are sent on offensive missions against us), let alone myriad alternative scenarios beyond the scope of human imagination. A deep learning artificial intelligence solution is proposed that will generate sufficient scenario data to allow UAS operators to train for situations that have not yet been encountered. Additionally, the proposed deep learning solution allows easy modifications of UAS geometries making it possible to test scenarios using alternative form factors for any UAS, or discovering design flaws in existing form factors that could hinder certain required rapid maneuvers. The proposed algorithm development would be carried out with an eye to future extensibility into real-world, rigid-body systems.