Artificial Intelligence Applications to Navy Avioni

THE ADDITION OF INCREASINGLY COMPLEX ACOUSTIC, NON-ACOUSTIC ASW, TARGET DETECTION, AND TRACKING EQUIPMENT ON NAVY HELICOPTERS HAS GREATLY INCREASED ASW CREW WORKLOAD. THE INCORPORATION OF ARTIFICIAL INTELLIGENCE TECHNIQUES CAN HELP REDUCE AIRCREW WORKLOAD IN THE AREA OF AIR, SURFACE, AND SUB-SURFACE THREATS. MISSION RESEARCH CORPORATION PROPOSES TO DESIGN A SYSTEM THAT PERFORMS THREAT CLASSIFICATION IN THREE DIFFERENT BUT HIGHLY COMPLEMENTARY WAYS: CLASSIFICATION BY THREAT IDENTITY, OPTICAL THREAT RESPONSE, AND THREAT PRIORITIZATION. THE SYSTEM USES AN ENTIRELY DATA DRIVEN APPROACH, ALLOWING FOR CHANGES IN THREAT AND COUNTERMEASURE SYSTEMS WITHOUT SYSTEM RECODING. IT UTILIZES EFFICIENT DATA COMPRESSION TECHNIQUES FOR THE MANAGEMENT OF THE LARGE AMOUNT OF DATA REQUIRED TO REPRESENT THREATS AND COUNTERMEASURES. THESYSTEM INCORPORATES AN INNOVATIVE TRAINABLE LINEAR OPTIMIZATION TECHNIQUE, THE MIMIC NET, FOR CLASSIFICATION BY RESPONSE AND PRIORITIZATION. SENSOR DATA MEASUREMENT UNCERTAINTY IS MITIGATED BY REPRESENTING MEASURED THREAT FEATURES IN TERMS OF FUZZY SETS. WE WILL ALSO EVALUATE OUR SYSTEM-S PERFORMANCE BY TESTING IT WITH A SET OF REPRESENTATIVE MULTI-THREAT ASW SCENARIOS.