• Journal of the Korea Institute of Military Science and Technology
• /
• v.26 no.1
• /
• pp.44-53
• /
• 2023

Adversarial attacks have received great attentions for their capacity to distract state-of-the-art neural networks by modifying objects in physical domain. Patch-based attack especially have got much attention for its optimization effectiveness and feasible adaptation to any objects to attack neural network-based object detectors. However, despite their strong attack performance, generated patches are strongly perceptible for humans, violating the fundamental assumption of adversarial examples. In this paper, we propose a camouflaged adversarial patch optimization method using military camouflage assessment metrics for naturalistic patch attacks. We also investigate camouflaged attack loss functions, applications of various camouflaged patches on army tank images, and validate the proposed approach with extensive experiments attacking Yolov5 detection model. Our methods produce more natural and realistic looking camouflaged patches while achieving competitive performance.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.47 no.2
• /
• pp.74-83
• /
• 2024

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.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.9
• /
• pp.836-842
• /
• 2015

In this paper, the target size and model based target size estimator (TMBE) algorithm is presented for iimaging infrared (IIR) seeker. At the imaging seeker, target size information is important factor for accurate tracking. The model based target size estimator filter (MBEF) algorithm was proposed to estimate target size at imaging infrared seeker. But, the model based target size estimator filter algorithm need to know relative distance from the target. In order to overcome the problem, we propose target size and model based target size estimator filter (TMBEF) algorithm which based on the target size. The performance of proposed algorithm is tested at target intercept scenario. The experiment results show that the proposed algorithm has the accurate target size estimating performance.

• Structural Engineering and Mechanics
• /
• v.46 no.5
• /
• pp.595-613
• /
• 2013

A new effective model for calculation of the equivalent uniform blast load for non-uniform blast load such as close-in explosion of a one-way square and rectangle reinforced concrete slab is proposed in this paper. The model is then validated using single degree of freedom (SDOF) system with the experiments and blast tests for square slabs and rectangle slabs. Test results showed that the model is accurate in predicting the damage level on the tested RC slabs under the given explosive charge weight and stand-off distance especially for close-in blast load. The results are also compared with those obtained by conventional SDOF analysis and finite element (FE) analysis using solid elements. It is shown that the new model is more accurate than the conventional SDOF analysis and is running faster than the FE analysis.

• Bulletin of the Korean Mathematical Society
• /
• v.53 no.5
• /
• pp.1327-1339
• /
• 2016

Self-organizing systems arise very naturally in artificial intelligence, and in physical, biological and social sciences. In this paper, we modify the classic Cucker-Smale model at both microscopic and macroscopic levels by taking the target motion pattern driving forces into consideration. Such target motion pattern driving force functions are properly defined for the line-shaped motion pattern and the ball-shaped motion pattern. For the modified Cucker-Smale model with the prescribed line-shaped motion pattern, we have analytically shown that there is a flocking pattern with an asymptotic flocking velocity. This is illustrated by numerical simulations using both symmetric and non-symmetric pairwise influence functions. For the modified Cucker-Smale model with the prescribed ball-shaped motion pattern, our simulations suggest that the solution also converges to the prescribed motion pattern.

• International Journal of Aeronautical and Space Sciences
• /
• v.16 no.4
• /
• pp.537-547
• /
• 2015

A Monte Carlo (MC) simulation was conducted to predict the reliability of a newly developed pyrotechnic pin puller. The reliability model is based on the stress-strength interference model that states that failure occurs if the stress exceeds the strength. In this study, the stress is considered to be the energy consumed by movement of a pin shaft, and the strength is considered to be the energy generated by pyrotechnic combustion for driving the pin shaft. Failure of the pin puller can thus be defined as the consumed energy being greater than the generated energy. These energies were calculated using a performance model formulated in the previous study of the present authors. The MC method was used to synthesize the probability densities of the two energies and evaluate the reliability of the pin puller. From a probabilistic perspective, the calculated reliability was compared to a deterministic safety factor. A sensitivity analysis was also conducted to determine which design parameters most affect the reliability.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.1
• /
• pp.31-50
• /
• 2023

In the midst of dynamic industrial changes, companies need data analysis considering the effects of integration of various technologies in order to establish innovative R & D strategies. However, the existing technology forecasting model evaluates individual technologies without considering relationship among them. To improve this problem, this study suggests a new methodology reflecting the integration of technologies. In the study, a technology forecasting indicator was developed using the technology integration index based on social network analysis. In order to verify the validity of the proposed methodology, 'drone task performance technology' based on patent data was applied to the research model. This study aimed to establish a theoretical basis to design a research model that reflects the degree of integration of technologies when conducting technology forecasting research. In addition, this study is meaningful in that it quantitatively verified the proposed methodology using actual patent data.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.47 no.2
• /
• pp.65-73
• /
• 2024

This paper presents a path planning optimization model for the engineering units to install obstacles in the shortest time during wartime. In a rapidly changing battlefield environment, engineering units operate various engineering obstacles to fix, bypass, and delay enemy maneuvers, and the success of the operation lies in efficiently planning the obstacle installation path in the shortest time. Existing studies have not reflected the existence of obstacle material storage that should be visited precedence before installing obstacles, and there is a problem that does not fit the reality of the operation in which the installation is continuously carried out on a regional basis. By presenting a Mixed Integrer Programming optimization model reflecting various constraints suitable for the battlefield environment, this study attempted to promote the efficient mission performance of the engineering unit during wartime.

• Journal of Digital Convergence
• /
• v.14 no.11
• /
• pp.81-90
• /
• 2016

In the civil market, companies launch new products when the acquired innovations are ready, but in defense, customer requests the innovation applied weapon systems. The technology adoption cycle model takes unusual form of market like inverse chasm takes technology inversely. This thesis describes an inverse chasm from the perspective of technology adoption cycle, equipped product model, and technical progress S-curve. As a way to overcome the inverse chasm, considering factors like a self-investment demo model, customer needs, and the temporary chasm expanding phenomenon are derived. And order-effective relationship analysis and chasm marketing strategy are suggested. Especially securing the core technologies and possibility for equipped product by developing self-investment demo model are identified as a good marketing strategy of chasm. This analysis and strategy suggest the policy implications for preemptive advantage of market positioning in the procurement process of defense, discontinuous innovation technology applied on.

• Journal of the Society of Disaster Information
• /
• v.19 no.1
• /
• pp.44-59
• /
• 2023

Purpose: In the event of mass casualties, triage must be done promptly and accurately so that as many patients as possible can be recovered and returned to the battlefield. However, medical personnel have received many tasks with less manpower, and the battlefield for classifying patients is too complex and uncertain. Therefore, we studied an artificial intelligence model that can assist and replace medical personnel on the battlefield. Method: The triage model is presented using reinforcement learning, a field of artificial intelligence. The learning of the model is conducted to find a policy that allows as many patients as possible to be treated, taking into account the condition of randomly set patients and the medical capability of the military hospital. Result: Whether the reinforcement learning model progressed well was confirmed through statistical graphs such as cumulative reward values. In addition, it was confirmed through the number of survivors whether the triage of the learned model was accurate. As a result of comparing the performance with the rule-based model, the reinforcement learning model was able to rescue 10% more patients than the rule-based model. Conclusion: Through this study, it was found that the triage model using reinforcement learning can be used as an alternative to assisting and replacing triage decision-making of medical personnel in the case of mass casualties.