AI-Driven, Secure Navy Mission Planning via Deep Reinforcement Learning and Attribute-Based Multi-Level Security

Current mission planning systems allow strike planners and operations centers to perform time-sensitive strike planning, execution monitoring, and validate mission effects using XML-based tools that visualize time critical attack plan and track plan status vs. execution. In this proposed STTR Phase I design for the Next Generation Navy Mission Planning (NGNMPS) system, we will identify expanded opportunities for the application of AI and ML algorithms/tools for intelligent, autonomous, and high-fidelity mission planning. The proposed AI approach can support NGNMPS by providing more accurate, less labor intensive, and increased fidelity strike planning. This proposed concept also takes a novel approach for the application of AI in the execution phase of the strike mission. By applying next generation ML processes for wireless communication systems, support for true digital interoperability and consequently enhanced performance across the Execution Mission Phase will be achieved. In addition to meeting the MLS and Cyber compliance requirements through the application of current methods and standards based procedures, the research aspect of this project will apply contemporary deep learning and other AI/ML algorithms to predict, identify, and counter the sophisticated attacks.