어뢰 기술 발전 동향 및 AI 활용 방안

황인혁^*

해군사관학교 조선공학과 부교수

Trends in Torpedo Technology Development and Application of Artificial Intelligence

Inhyuck Hwang^*

Associate professor, Dept. of Naval Architecture, Republic of Korea Naval Academy

Correspondence to: *Inhyuck Hwang Dept. of Naval Architecture, ROK Naval Academy 1 Jungwon-ro, Jinhae-gu, Changwon-si, Gyungsangnam-do, 51704, Republic of Korea Tel: +82-55-907-5322 E-mail: ih.hwang@navy.ac.kr

초록

본 연구는 어뢰 기술을 다섯 분야로 분류하고, 각 기술군에서 AI 기술을 적용할 수 있는 가능성을 분석하였다. 음향 기반 표적 인식과 강화학습 기반 기동 및 회피 분야에서 AI 기술 적용 접합도가 높은 것으로 판단하였다. 실제 연구 사례에서도 예상한 두가지 분야의 AI 활용 연구가 많이 이루어지고 있었다. 또한 군사 데이터 접근 제한과 검증체계 부재로 인해 AI 기술의 적용에는 구조적 한계를 확인하고 본 연구는 데이터 확보·활용·검증을 통합 지원하는 국방 AI 데이터센터 구축을 제안하여 향후 AI 기반의 무기체계 발전을 위한 기반 마련의 필요성을 강조하였다.

Abstract

This study classifies torpedo technologies into five domains and evaluates the applicability of artificial intelligence (AI) to each. AI shows the highest suitability in acoustic target recognition and reinforcement learning based maneuver and evasion, which aligns with recent research trends focused on these areas. The study also identifies structural limitations restricted access to military data and the lack of verification frameworks that hinder practical AI adoption. To address these issues, the establishment of a National Defense AI Data Center is proposed as essential infrastructure for future AI based weapon system development.

Keywords:

Torpedo, Artificial Intelligence, Acoustic Target Recognition, Reinforcement Learning, Defense AI Data Center

키워드:

어뢰, 인공지능, 음향표적인식, 강화학습, 국방 AI 데이터 센터

Acknowledgments

본 논문은 해군사관학교 해양연구소의 연구비 지원을 받아 수행한 연구결과임(과제명: 어뢰 기만 기술 동향 및 발전 방향).

References

Min-Ki Jang, “A Study on Improvement of Storage Safety through Quality improvement of Torpedo Propulsion Battery.” Journal of Korea Academia-Industrial Cooperation Society, 2019, 20(7), pp. 291-298.
Ki-Hun Kim, Hyun-Taek Choi, Chong-Moo Lee, Sea-Moon Kim, Pan-Mook Lee & Seong-Kwon Cho, “Implementation of Deep-sea UUV Precise Underwater Navigation Based on Multiple Sensor Fusion.” Journal of Ocean Engineering and Technology, 2010, 24(3), pp. 46-51.
Young-Hyun Lee, Bon-Hwa Ku, Suk-Moon Chung, Woo-Young Hong and Han-Seok Ko, “Robust Search Method for Ship Wake Using Two Wake Sensors.” The Journal of the Acoustical Society of Korea, 2010, 29(3), pp.155-164.
J.E. Kye, J.I. Cho, W.P. Yoo, S.L. Choi & J.H. Park, “Trends and Applications on Multi-beam Side Scan Sonar Sensor Technology.” Electronics and Telecommunications Trends, 2013, 28(6), pp. 167-179.
J.W. Choi, S.H. Kim & S.W. Son, “Characteristics of Underwater Acoustic Wireless Communication Channels according to Marine Environmental Fluctuations.” Information and Communications Magazine, 2016, 33(8), pp.52-62.
U.S. Kim, H.-Y. Jeong & K.-M. Lee, “Development of Surface Ship Torpedo Defense System Simulator Considering Bubble-Generating Wake Decoys.” Journal of the Korean Society of Military Science and Technology, 2024, 27(3), pp. 416-427.
M.I. Shin, W.Y. Hong, J.H. Lee, “Effectiveness Analysis of a Hard-kill Underwater Defense System for Surface Warships against Wake-homing Torpedo Attack.” Journal of Advances in Military Studies, 2023, 6(2) pp.1-15.
K.S. Park, B.M. Yoon, J.H. Lee & K.C. Shin, “Ensemble Design of Machine Learning Techniques: Experimental Validation by Prediction of Feature Information of Torpedo Targets.” KIISE Transactions on Computing Practices, 2021, 27(6), pp. 273–281.
J.H. Chung, G.S. Kim, S.H. Park, J.H. Kim, W.H. Yun, “Reinforcement Learning-based Counter Measure Tactics to Avoid Torpedo Threats.” The Journal of Korean Institute of Communications and Information Science, 2024, 49(3), pp. 333-345.
E.J. Roh, H.S. Lee, S.H. Park, J.H. Kim, K.H. Kim & S.H. Kim, “Directional Autonomous Torpedo Maneuver Control Using Reinforcement Learning.” The Journal of the Korean Institute of Communications and Information Sciences, 2024, 49(5), pp. 752–761.
Jianjing Deng, Xiangfeng Yang, Liwen Liu, Lei Shi, Yongsheng Li and Yunchuan Yang, “Real-Time Underwater Acoustic Homing Weapon Target Recognition Based on a Stacking Technique of Ensemble Learning.” Journal of Marine Science and Engineering, 2023, 11(12), 2305.
Kunchul Hwang & Jinwhan Kim, “Wake Homing Torpedo Guidance Using a Hierarchical Deep Reinforcement Learning Framework.” IEEE Access, 2025, 13, pp.72938-72952.
Zhong Wang, Zhiwen Wen, Weitong Cui, Daming Zhou and Pei Wang, “Design of Reinforcement Learning Guidance Law for Antitorpedo Torpedoes.” International Journal of Aerospace Engineering. 2025. [https://doi.org/10.1155/ijae/5146939]