AlphaMosaic: An Artificially Intelligent Battle Management Architecture
- 1 March 2022
- journal article
- research article
- Published by American Institute of Aeronautics and Astronautics (AIAA) in Journal of Aerospace Information Systems
- Vol. 19 (3), 203-213
- https://doi.org/10.2514/1.i010991
Abstract
Warfare is increasing in complexity, speed, and scale—not only due to enhanced technological capabilities but also from the employment methodologies associated with them. Incorporating artificial intelligence (AI) technology into this realm is a cogent solution to help address these complications because of the reduced cost, reduced risk to human life, and increased capability to rapidly adapt to changing environments. However, the introduction of AI comes with a host of new considerations. If AI is to be successfully integrated into air combat, humans must be included in the AI processing loop, and human interaction with AI decision loops must be frictionless. Additionally, AI-supported battle management systems must be designed for high and increasing human trust across dynamically changing scenarios. This paper presents AlphaMosaic, an AI battle manager developed as part of the Defense Advanced Research Projects Agency Air Combat Evolution program that is designed to incorporate human feedback in a manner conducive to true manned–unmanned aircraft teaming in beyond visual range air-combat scenarios.Keywords
Funding Information
- Defense Advanced Research Projects Agency (HRC001120C0061)
This publication has 6 references indexed in Scilit:
- Developing a Digital Twin and Digital Thread Framework for an ‘Industry 4.0’ ShipyardApplied Sciences, 2021
- Multi-Agent Systems: A SurveyIEEE Access, 2018
- StarCraft Micromanagement With Reinforcement Learning and Curriculum Transfer LearningIEEE Transactions on Emerging Topics in Computational Intelligence, 2018
- DeepStack: Expert-level artificial intelligence in heads-up no-limit pokerScience, 2017
- Mastering the game of Go with deep neural networks and tree searchNature, 2016
- Fuego—An Open-Source Framework for Board Games and Go Engine Based on Monte Carlo Tree SearchIEEE Transactions on Computational Intelligence and AI in Games, 2010