Summaries - Office of Research & Innovation
Research Summaries
Back Applying Reinforcement Learning to Predict Adversary's Response in a Naval Conflict
Fiscal Year | 2023 |
Division | Research & Sponsored Programs |
Department | Naval Research Program |
Investigator(s) | Barton, Armon C. |
Sponsor | NPS Naval Research Program (Navy) |
Summary | As the US Navy and its allies maintain presence around the world to ensure freedom of the seas, it is increasingly necessary to identify the optimal strategy in engagements with other vessels. The expectation is that the captains of our warships can correctly identify the optimal moves for every interaction. We have seen examples that show the contrary: namely the USS John McCain and USS Fitzgerald. With the advancement of Artificial Intelligence (AI) over the years, scientists have shown that they are able to implement Reinforcement Learning (RL) agents to defeat world class professionals in games like Chess, Poker, StarCraft and more. The process of creating a tool that leverages RL techniques in naval scenarios could help provide a decision aid to ship captains to identify the optimal move in a given scenario. This work will implement a simulation-based program that can take naval scenario-based environments with RL techniques and demonstrate the ability to identify optimal movements for both own forces and opposing. This would provide a backup to ship captains when deciding what actions to take in their day-to-day operations. |
Keywords | Reinforcement, learning, adversary, response, tracking, prediction |
Publications | Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal |
Data | Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal |