멀티에이전트 기반 Deep Q-Network 모델을 이용한 동적 미사일 방어효과 개선
Improving Dynamic Missile Defense Effectiveness Using Multi-Agent Deep Q-Network Model
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
The threat of North Korea's long-range firepower is recognized as a typical asymmetric threat, and South Korea is prioritizing the development of a Korean-style missile defense system to defend against it. To address this, previous research modeled North Korean long-range artillery attacks as a Markov Decision Process (MDP) and used Approximate Dynamic Programming as an algorithm for missile defense, but due to its limitations, there is an intention to apply deep reinforcement learning techniques that incorporate deep learning. In this paper, we aim to develop a missile defense system algorithm by applying a modified DQN with multi-agent-based deep reinforcement learning techniques. Through this, we have researched to ensure an efficient missile defense system can be implemented considering the style of attacks in recent wars, such as how effectively it can respond to enemy missile attacks, and have proven that the results learned through deep reinforcement learning show superior outcomes.
Abstract
The threat of North Korea's long-range firepower is recognized as a typical asymmetric threat, and South Korea is prioritizing the development of a Korean-style missile defense system to defend against it. To address this, previous research modeled North Korean long-range artillery attacks as a Markov Decision Process (MDP) and used Approximate Dynamic Programming as an algorithm for missile defense, but due to its limitations, there is an intention to apply deep reinforcement learning techniques that incorporate deep learning. In this paper, we aim to develop a missile defense system algorithm by applying a modified DQN with multi-agent-based deep reinforcement learning techniques. Through this, we have researched to ensure an efficient missile defense system can be implemented considering the style of attacks in recent wars, such as how effectively it can respond to enemy missile attacks, and have proven that the results learned through deep reinforcement learning show superior outcomes.
참고문헌 (30)
- Bertsekas, D., "Missile Defense and Interceptor Allocation by Neuro-Dynamic Programming", IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans, 30(1), pp.42~51 (2000)
- Cha, Y. H., "Exact Algorithm for the Weapon Target Assignment and Fire Scheduling Problem", Journal of the Society of Korea Industrial and Systems Engineering, 42(1), pp.143~150 (2019)
- Davis, M. T., "Approximate Dynamic Programming for Missile Defense Interceptor Fire Control", European Journal of Operational Research, 259, pp.873~886 (2017)
- Im, J. S., "A Study of Multi-to-Majority Response on Threat Assessment and Weapon Assignment Algorithm : by Adjusting Ballistic Missiles and Long-Range Artillery Threat", Journal of Korean Society of Industrial and Systems Engineering, 44(4), pp.43~52 (2021)
- Jang, B. C., "Consideration on Our Asymmetric Response through the Israel-Hamas Surprise Attack", Defense & Technology, 538, pp.116~125 (2023)
- Jang, J. G., "A Linear Approximation Model for an Asset-based Weapon Target Assignment Problem", Journal Society of Korea Industrial and System Engineering, 38(3), pp.108~116 (2015)
- Jung, J. K., "Rolling-Horizon Scheduling Algorithm for Dynamic Weapon-Target Assignment in Air Defense Engagement", Journal of the Korean Institute of Industrial Engineering, 46(1), pp.11~24 (2020)
- Kim, H. H., "Reinforcement Learning-based Dynamic Weapon Allocation for Multiple Long-range Artillery Attacks", Journal of the Korean Institute of Industrial Management Systems, 45(4), pp.42~52 (2022)
- Kim, J. H., "An Application of Quantum-inspired Genetic Algorithm for Weapon Target Assignment Problem", Journal Society of Korea Industrial and System Engineering, 40(4), pp.260~267 (2017)
- Lee, C. S., "Approximate Dynamic Programming Based Interceptor Fire Control and Effectiveness Analysis for M-To-M Engagement", Journal of the Korean Society for Aeronautical & Space Sciences, 50(4), pp.287~295 (2022)
- Lee, W. W., "Reinforcement Learning with Python and Keras", pp.227~247 (2020)
- Lee, Z. J., "An Immunity Based Ant Colony Optimization Algorithm for Solving Weapon-Target Assignment Problem", Applied Soft Computing, 2(1), pp.39~47 (2002)
- Li, S. E., "Deep Reinforcement Learning, Reinforcement Learning for Sequential Decision and Optimal Control", pp.365~402 (2023), Springer Nature Singapore
- Mnih, V., "Human-level Control Through Deep Reinforcement Learning", Nature, 518(7540), pp.529~533 (2015)
- Naeem, H., "An Optimal Dynamic Threat Evaluation and Weapon Scheduling Technique", Know-ledge-Based Systems, 23(4), pp.337~342 (2010)
- Park, Y.W., "Formulation of a Defense Artificial Intelligence Development Plan", Korean Society for Defense Technology, pp.3~8 (2020)
- Powell, W. B., "Approximate Dynamic Programming:Solving the Curse of Dimensionality", pp.315~346 (2011), John Wiley & Sons
- Powell, W. B., "Approximate Dynamic Programming:Solving the Curse of Dimensionality", pp.235~276 (2011), John Wiley & Sons
- Powell, W. B., "Perspectives of Approximate Dynamic Programming", Annals of Operations Research, 13(2), pp.1~38 (2012)
- Schulman, J., "Trust Region Policy Optimization", Proceedings of The 32nd International Conference on Machine Learning, pp.1~9 (2015)
- "segye news"[URL]
- Shin, M. K., "Mean Field Game based Reinforcement Learning for Weapon-Target Assignment", Journal of the Korea Institute of Military Science and Technology, 23(4), pp.337~345 (2020)
- Summers, D. S., "An Approximate Dynamic Programming for Comparing Firing Policies in a Networked Air Defense Environment", Computers & Operations Research, 117, pp.1~29 (2020)
- Sutton, R. S., "Reinforcement learning:An introduction", pp.30~39 (2018)
- "Tutorials for Reinforcement Learning"[URL]
- "Yonhapnews"[URL]
- "Yonhapnews"[URL]