• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.2
• /
• pp.256-264
• /
• 2024

In the last 10 years, AI has made rapid progress, and image classification, in particular, are showing excellent performance based on deep learning. Nevertheless, due to the nature of deep learning represented by a black box, it is difficult to actually use it in critical decision-making situations such as national defense, autonomous driving, medical care, and finance due to the lack of explainability of judgement results. In order to overcome these limitations, in this study, a model description algorithm capable of local interpretation was applied to the inception network-derived AI to analyze what grounds they made when classifying national defense data. Specifically, we conduct a comparative analysis of explainability based on confidence values by performing LIME analysis from the Inception v2_resnet model and verify the similarity between human interpretations and LIME explanations. Furthermore, by comparing the LIME explanation results through the Top1 output results for Inception v3, Inception v2_resnet, and Xception models, we confirm the feasibility of comparing the efficiency and availability of deep learning networks using XAI.

• Journal of the Korean Society of Systems Engineering
• /
• v.19 no.1
• /
• pp.91-106
• /
• 2023

Multidimensional efforts such as budgeting, organizing, and institutionalizing are being carried out for the adoption of defense AI. However, there is little interest in eliminating duplication of defense resources that may occur during the AI adoption. In this study, we propose a theoretical conceptual model to optimize duplication of AI technology that may occur during the AI adoption in the vast defense field. For a systematic approach, the JCA of the US DoD and system abstraction method are applied, and the IMO logical structure is used to decompose AI requirements and identify duplication. As a result of analyzing the effectiveness of our conceptual model through six example defense AI requirements, it was found that the amount of requirements of data and AI technologies could be reduced by up to 41.7% and 70%, respectively, and estimated costs could be reduced by up to 35.5%.

• Convergence Security Journal
• /
• v.22 no.5
• /
• pp.135-143
• /
• 2022

The defense environment is rapidly changing, such as nuclear and missile threats of North Korea, changes in war patterns, and a decrease in military service resources due to low birth rate. In order to actively respond to these changes, the Korean military is promoting Defense Innovation 4.0 and is trying to foster an army armed with high technology such as artificial intelligence (AI), big data analysis, etc. In this regard, we analyze the effectiveness of the radar-based AI scientific guard system applied by high technology for guard operations using Analytic Hierarchy Process (AHP). We first select evaluation factors that can assess the effectiveness of the scientific guard system, and analyze its relative importance. Each evaluation factor was selected by deriving a significant concept from operating principle and how they work, and by consulting experts on the correlation between each factor and effectiveness of the scientific guard system. We examine the relative effects of the radar-based AI scientific guard system and existing scientific guard system based on the importance of the evaluation factors.

• Journal of Software Assessment and Valuation
• /
• v.17 no.2
• /
• pp.59-73
• /
• 2021

Recently, artificial intelligence has spread rapidly and popularized and expanded to the voice recognition personal service sector, and major countries have established artificial intelligence promotion strategies, but in the case of South Korea's defense domain, its influence is low with a geopolitical location with North Korea. This paper presents a total of six strategies for promoting South Korea's defense artificial intelligence, including establishing roadmaps, securing manpower, installing the artificial intelligence base, and strengthening cooperation among stakeholders in order to increase the impact of South Korea's defense artificial intelligence and successfully promote artificial intelligence. These suggestions are expected to establish the foundation for expanding the base of artificial intelligence.

• Defense and Technology
• /
• no.10 s.92
• /
• pp.11-17
• /
• 1986

• Journal of The Korean Institute of Defense Technology
• /
• v.5 no.2
• /
• pp.1-9
• /
• 2023

The purpose of this study is to derive the need to improve the test and evaluation(T&E) performance system of the weapon systems to which advanced science and technology is applied, evaluate priorities, and present development plans. T&E of Al-based weapon systems through a literature research and case analysis on changes in the T&E environment for weapon system, 11 detailed evaluation items were derived from the in terms of the T&E system, structure, and technology. Al-based weapon system test evaluation should be performed in parallel with data-based performance evaluation and actual T&E, and performance measurement using a separate T&E data set is required for AI models performance evaluation. As a result of analyzing the importance of T&E through AHP analysis, the order of T&E system-technology-structure was evaluated, and the priority of detailed evaluation items was evaluated in the order of T&E result judgment-T&E organization and expert training-scientific T&E. For evaluation items with high priority, measures to improve the T&E performance system were presented.

• Journal of the Korean Society of Systems Engineering
• /
• v.20 no.1
• /
• pp.64-75
• /
• 2024

With the development of the Internet, the information in our lives has become fast and diverse. Especially in the field of defense, articles and information are pouring in from various sources every day, and fast information selection, understanding, and decision-making are required in the ever-changing situation. It is very cumbersome to go from platform to platform and read articles one by one to get the information you need. To solve this problem, this research aims to save time and provide quick access to the latest information by allowing you to quickly grasp key information from summarized content without having to read the entire article. This can improve efficiency by allowing defense professionals to focus more on important tasks rather than extensive information search and analysis.

• Journal of the military operations research society of Korea
• /
• v.22 no.1
• /
• pp.67-80
• /
• 1996

Artificial Intelligence(AI) techniques and Object-Oriented(OO) techniques contribute to the simulation modeling of the complex systems. AI techniques are suitable to model human reasoning in the simulation. While OO techniques have advantages of re-usability, maintainability and extendability of the software. Thus, in this paper, we design a knowledge-based object-oriented simulation model, particularly for the logistics analysis of military armor vehicles. The simulation model consists of three modules i.e., scenario, simulation mechanism, and inference engine. The model is designed within the OO paradigm and implemented by using the C++ language. An example case of using the model for the logistic analysis is included.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.5
• /
• pp.567-580
• /
• 2024

Due to specificity in the underwater environment, the difficulty of detecting submarine and the threat of submarine are increasing. The probability of detecting a submarine can be increased by estimation submarine path and optimizing sonobuoy deployment. In this paper, marine data collection, dynamics of submarine, submarine tracking path modeling, acoustic wave propagation modeling, detection probability modeling are applied in the simulator as similar to reality as possible. A simulator is developed to design submarine path estimation and sonobuoy deployment optimization scenario and to check result according to the scenario.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.51-54
• /
• 2024

본 논문에서는 온 디바이스 국방 AI를 위한 효율적인 학습 방법을 제안한다. 제안하는 방법은 모델 전체를 재학습하는 대신 필요한 부분만 세밀하게 조정하여 계산 비용과 시간을 대폭 줄이는 PEFT 기법의 LoRa를 적용하였다. LoRa는 기존의 신경망 가중치를 직접 수정하지 않고 추가적인 낮은 랭크의 매트릭스를 학습하는 방식으로 기존 모델의 구조를 크게 변경하지 않으면서도, 효율적으로 새로운 작업에 적응할 수 있다. 또한 학습 파라미터 및 연산 입출력에 데이터에 대하여 32비트의 부동소수점(FP32) 대신 부동소수점(FP16, FP8) 또는 정수형(INT8)을 활용하는 경량화 기법인 양자화도 적용하였다. 적용 결과 학습시 요구되는 GPU의 사용량이 32GB에서 5.7GB로 82.19% 감소함을 확인하였다. 동일한 조건에서 동일한 데이터로 모델의 성능을 평가한 결과 동일 학습 횟수에선 LoRa와 양자화가 적용된 모델의 오류가 기본 모델보다 53.34% 증가함을 확인하였다. 모델 성능의 감소를 줄이기 위해서는 학습 횟수를 더 증가시킨 결과 오류 증가율이 29.29%로 동일 학습 횟수보다 더 줄어듬을 확인하였다.