• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.1
• /
• pp.31-42
• /
• 2024

In this study, the methodology of determination of base bleed motor data for base bleed projectile based on the NATO standard trajectory model, especially STANAG 4355 Method 2 were presented. Ground combustion experiments and aerodynamic performance firing tests were conducted to determine the drag reduction motor data of the base bleed projectile and this data was described based on the NATO standard ballistic model. The derived drag reduction motor data were input into the ballistic equations to complete the ballistic model and it was confirmed that the calculated predicted trajectory from the ballistic model matched well with the measured trajectory from the aerodynamic performance firing tests.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.3
• /
• pp.346-354
• /
• 2024

Deep learning models require high computing power due to a substantial amount of computation. It is difficult to use them in devices with limited computing environments, such as coastal surveillance equipments. In this study, a lightweight model is constructed by analyzing the weight changes of the convolutional layers during the training process based on MobileNet and then pruning the layers that affects the model less. The performance comparison results show that the lightweight model maintains performance while reducing computational load, parameters, model size, and data processing speed. As a result of this study, an effective pruning method for constructing lightweight deep learning models and the possibility of using equipment resources efficiently through lightweight models in limited computing environments such as coastal surveillance equipments are presented.

• Journal of the Korean Society of Combustion
• /
• v.22 no.2
• /
• pp.36-41
• /
• 2017

Aero gas turbine engines must demonstrate their ability to be ignited on ground conditions or relighted in flight. The electric spark ignition is usually used in current aero gas turbine engines. Experiments on ignition characteristics relating to spark igniter penetration depth under atmospheric pressure and temperature conditions were conducted on the model combustor which is scaled in 1/18. Exciter was operated during 2 seconds, and successful ignition phenomena were confirmed by the pressure rising sharply in combustor. In addition, instantaneous ignition images were captured by a high-speed camera. It showed kernel propagation and successful ignition events in the sector model combustor. Ignition test results showed that ignition limit with increase in penetration depth of the igniter plug was wider. When the penetration depth of the igniter plug increased under the same fuel injection pressure condition, successful ignition events were obtained in higher differential pressure conditions between inlet and outlet of the combustor. The results demonstrate that the ratio of the combustible mixture, which is exposed to the high temperature environment around the igniter plug tip, increases. Thereby affect the combustor ignition performance.

• Journal of the Korea Society for Simulation
• /
• v.29 no.2
• /
• pp.49-61
• /
• 2020

It is important to operate a limited number of interceptors effectively to counter ballistic missile threats. The existing interceptor operating method determines the number of interceptors according to the level of TBM (Theater Ballistic Missile) engagement effectiveness applied to a defended asset. It can cause either excessive interceptor waste compared to the intercept probability or the intercept probability decrease. Thus, interceptor operating method must be decided considering the number of ballistic missiles, intercept probability and cost. This study proposes a mathematical model to improve the existing interceptor operating method. In addition, the efficiency indicator is proposed for trade-off between intercept probability and cost. As a result of the simulations, the mathematical model-based interceptor operating method can achieve better results than the existing interceptor operating method.

• The KIPS Transactions:PartD
• /
• v.12D no.5 s.101
• /
• pp.729-736
• /
• 2005

MND(Ministry of National Defense) has developed MND AF(Ministry of National Defense Architecture Framework) and CADM(Core Architecture Data Model) to guarantee interoperability among defense information systems. TRM(Technical Reference Model) and SP(Standard Profile) product defined in MND AF is core part of Information Technology Architecture and core element of interoperability guarantee. In this paper, we proposed a method which manages technical service and standard of TRM and SP, and developed TRM and SP management system based on the method. TRM and SP management system provides the basis for interoperability among information systems and a more efficient development and management of TRM and SP product.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.24 no.4
• /
• pp.382-391
• /
• 2021

Unsupervised domain adaptation often gives impressive solutions to handle domain shift of data. Most of current approaches assume that unlabeled target data to train is abundant. This assumption is not always true in practices. To tackle this issue, we propose a general solution to solve the domain gap minimization problem without any target data. Our method consists of two regularization steps. The first step is a pixel regularization by arbitrary style transfer. Recently, some methods bring style transfer algorithms to domain adaptation and domain generalization process. They use style transfer algorithms to remove texture bias in source domain data. We also use style transfer algorithms for removing texture bias, but our method depends on neither domain adaptation nor domain generalization paradigm. The second regularization step is a feature regularization by feature alignment. Adding a feature alignment loss term to the model loss, the model learns domain invariant representation more efficiently. We evaluate our regularization methods from several experiments both on small dataset and large dataset. From the experiments, we show that our model can learn domain invariant representation as much as unsupervised domain adaptation methods.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.13 no.1
• /
• pp.99-104
• /
• 2010

MDPREF(Multidimensional Preference Analysis) is a program for analysis of preferences. It is what is known as a vector model. This means that the objective of the MDPREF analysis is to identify a perceptual map displaying subject(attribute) vectors. To form the subject vectors visually, lines are drawn from the origin of the plot to each subject point. We analysis the defense industry infrastructure in fire power area by using MDPREF.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.4
• /
• pp.474-484
• /
• 2024

Reinforcement learning, which are also studied in the field of defense, face the problem of sample efficiency, which requires a large amount of data to train. Transfer learning has been introduced to address this problem, but its effectiveness is sometimes marginal because the model does not effectively leverage prior knowledge. In this study, we propose a stochastic initial state randomization(SISR) method to enable robust knowledge transfer that promote generalized and sufficient knowledge transfer. We developed a simulation environment involving a cooperative robot transportation task. Experimental results show that successful tasks are achieved when SISR is applied, while tasks fail when SISR is not applied. We also analyzed how the amount of state information collected by the agents changes with the application of SISR.

• Journal of National Security and Military Science
• /
• s.2
• /
• pp.261-283
• /
• 2004

To build up the military strength based on information oriented armed forces, the Korean ministry of national defense (MND) promotes the defense CALS (Continuous Acquisition and Life cycle Support) initiative for the reductions of acquisition times, improvements of system qualities, and reductions of costs. These defense CALS activities are the major component of the underlying mid and long term defense digitization program and the ultimate goal of program is to bring a quick victory by providing real-time battlefield intelligence and the economical operations of the military. The concept of defense CALS is to automate the acquisition and disposition of defense systems throughout their life cycle. For implementing defense CALS, the technology for exchange and sharing CALS standard data that is created once and used many times should be considered. In order to develop an efficient CALS information exchange and sharing system, it is required to integrate distributed and heterogeneous data sources and provide systematic search tools for those data. In this study, we developed a content based search engine technology which is essential for the construction of integrated data environments. The developed technology provides the environment of sharing the CALS standard data such as SGML(Standard Generalized Markup Language) and STEP(Standard for The Exchange of Product model data). Utilizing this technology, users can find and access distributed and heterogeneous data sources without knowing its actual location.

• Defense and Technology
• /
• no.4 s.242
• /
• pp.32-43
• /
• 1999

표준/극한대기 모델은 항공우주 분야에 다양하게 적용되고 있고, 특히 표준대기는 항공규정 정립 및 항공기개발 분야의 풍동실험과 컴퓨터 시뮬레이션 등을 위한 대기관련 표준데이터로 활용되는 것이며, 시시각각으로 변화하는 실제의 대기현상과는 상이한 가상의 표준(Standard)과 같으므로 실질적으로 극복해야 하는 운용여건을 판단하기 위해서는 극한대기 모델을 고려해야 할 것으로 본다