• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.18-20
• /
• 2021

In order to prevent collision accidents of ships, which has been increasing recently, research on artificial intelligence-based autonomously operated ships (Maritime Autonomous Surface Ship, MASS) is underway. However, most of the studies related to autonomous ships mainly target medium-to-large ships due to the size and cost of the autonomous navigation system, and the sensors used here have a problem in that it is difficult to mount them on small ships. Therefore, this paper provides a path tracking system equipped with GPS and IMU sensors for autonomous operation of small ships. GPS and IMU sensors are utilized to determine the exact position of the vessel, which allows the proposed system to manually control the small vessel model to create a path and then when the small vessel travels the same path. Use the Pure Pursuit algorithm to follow the path. As a result, In this research, it is expected that a lightweight and low-cost sensor can be used to develop an autonomous operation system for small ships at low cost.

• Journal of Advanced Navigation Technology
• /
• v.26 no.6
• /
• pp.464-473
• /
• 2022

In this paper, about a lightweight design that satisfies the performance of UHF-GPS Antenna used in autonomous underwater vehicle is proposed. Structural analysis, watertight external pressure test and non-destructive testing used in the design process are decided in consideration of structural safety for operating external forces in the underwater environment. First, the material of radome is selected for the performance of the UHF-GPS Antenna for communication with the carrier on the underwater operation in consideration of the 20 bar pressure generated. And the material of radome as PA-GF is selected by conducting electromagnetic field analysis and structural analysis and by considering high strength, rigidity and high dielectric constant. Electromagnetic field analysis and structural analysis by the thickness of radome are additionally performed in order to satisfy the required weight of UHF-GPS antenna. After selecting the final model, its structural safety is verified through watertight external pressure test and non-destructive testing.

• Journal of Biomedical Engineering Research
• /
• v.44 no.6
• /
• pp.392-403
• /
• 2023

This study presents a new hybrid fNIRS-EEG system to meet the demand for a lightweight and low-cost sleep pattern monitoring device. For multiple-channel configuration, a six-channel electroencephalogram (EEG) and a functional near-infrared spectroscopy (fNIRS) system with eight photodiodes (PD) and four dual-wavelength LEDs are designed. To enhance the convenience of signal measurement, the device is miniaturized into a patch-like form, enabling simultaneous measurement on the forehead. Due to its fully integrated functionality, the developed system is advantageous for performing sleep stage classification with high-temporal and spatial resolution data. This can be realized by utilizing a two-dimensional (2D) brain activation map based on the concentration changes in oxyhemoglobin and deoxyhemoglobin during sleep stage transitions. For the system verification, the phantom model with known optical properties was tested at first, and then the sleep experiment for a human subject was conducted. The experimental results show that the developed system qualifies as a portable hybrid fNIRS-EEG sleep pattern monitoring device.

• Advances in aircraft and spacecraft science
• /
• v.10 no.3
• /
• pp.223-243
• /
• 2023

There is a burgeoning demand for minimizing the mass of satellites because of its direct impact on reducing launch-to-orbit cost. This must be done without compromising the structure's efficiency. The present paper introduces a relatively low-cost and easily implementable approach for optimizing structural mass to a maximum natural frequency. The natural frequencies of the satellite are of utmost pertinence to the application requirements, as the sensitive electronic instrumentation and onboard computers should not be affected by the vibrations of the satellite structure. This methodology is applied to a realistic model of Al-Azhar University micro-satellite in partnership with the Egyptian Space Agency. The procedure used in structural design can be summarized in two steps. The first step is to select the most favorable primary structural configuration among several different candidate variants. The nominated variant is selected as the one scoring maximum relative dynamic stiffness. The second step is to use perforation patterns reduce the overall mass of structural elements in the selected variant without changing the weight. The results of the presented procedure demonstrate that the mass reduction percentage was found to be 39% when compared to the unperforated configuration that had the same plate thickness. The findings of this study challenge the commonly accepted notion that isogrid perforations are the most effective means of achieving the goal of reducing mass while maintaining stiffness. Rather, the study highlights the potential benefits of exploring a wider range of perforation unit cells during the design process. The study revealed that rectangular perforation patterns had the lowest efficiency in terms of modal stiffness, while triangular patterns resulted in the highest efficiency. These results suggest that there may be significant gains to be made by considering a broader range of perforation shapes and configurations in the design of lightweight structures.

• Journal of the Korean Society of Safety
• /
• v.39 no.2
• /
• pp.38-43
• /
• 2024

Recently, with the expanding market for electronic devices and electric vehicles, secondary battery usage has been on the rise. Lithium-ion batteries are particularly popular due to their fast charging times and lightweight nature compared to other types of batteries. A secondary battery consists of four components: anode, cathode, electrolyte, and separator. Generally, the positive and negative electrode materials of secondary batteries are composed of an active material, a binder, and a conductive material. Acetylene Black (AB) is utilized to enhance conductivity between active material particles or metal dust collectors, preventing the binder from acting as an insulator. However, when recycling waste batteries that have been subject to high usage, there is a risk of fire and explosion accidents, as accurately identifying the characteristics of Acetylene Black dust proves to be challenging. In this study, the lower explosion limit for Acetylene Black dust with an average particle size of 0.042 ㎛ was determined to be 153.64 mg/L using a Hartmann-type dust explosion device. Notably, the dust did not explode at values below 168 mg, rendering the lower explosion limit calculation unfeasible. Analysis of explosion delay times with varying electrode gaps revealed the shortest delay time at 3 mm, with a noticeable increase in delay times for gaps of 4 mm or greater. The findings offer fundamental data for fire and explosion prevention measures in Acetylene Black waste recycling processes via a predictive model for lower explosion limits and ignition delay time.

• Journal of The Korean Institute of Defense Technology
• /
• v.5 no.4
• /
• pp.16-21
• /
• 2023

Although the global number of landmines is on a declining trend, the damages caused by previously buried landmines persist. In light of this, the present study contemplates solutions to issues and constraints that may arise due to the improvement of mine detection equipment and the reduction in the number of future soldiers. Current mine detectors lack data storage capabilities, posing limitations on data collection for research purposes. Additionally, practical data collection in real-world environments demands substantial time and manpower. Therefore, in this study, gprMax simulation was utilized to generate data. The lightweight CNN-based model, MobileNet, was trained and validated with real data, achieving a high identification rate of 97.35%. Consequently, the potential integration of technologies such as deep learning and simulation into geographical detection equipment is highlighted, offering a pathway to address potential future challenges. The study aims to somewhat alleviate these issues and anticipates contributing to the development of our military capabilities in becoming a future scientific and technological force.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.249-263
• /
• 2023

Collaborative Filtering, a representative recommendation system methodology, consists of two approaches: neighbor methods and latent factor models. Among these, the latent factor model using matrix factorization decomposes the user-item interaction matrix into two lower-dimensional rectangular matrices, predicting the item's rating through the product of these matrices. Due to the factor vectors inferred from rating patterns capturing user and item characteristics, this method is superior in scalability, accuracy, and flexibility compared to neighbor-based methods. However, it has a fundamental drawback: the need to reflect the diversity of preferences of different individuals for items with no ratings. This limitation leads to repetitive and inaccurate recommendations. The Adaptive Deep Latent Factor Model (ADLFM) was developed to address this issue. This model adaptively learns the preferences for each item by using the item description, which provides a detailed summary and explanation of the item. ADLFM takes in item description as input, calculates latent vectors of the user and item, and presents a method that can reflect personal diversity using an attention score. However, due to the requirement of a dataset that includes item descriptions, the domain that can apply ADLFM is limited, resulting in generalization limitations. This study proposes a Generalized Adaptive Deep Latent Factor Recommendation Model, G-ADLFRM, to improve the limitations of ADLFM. Firstly, we use item ID, commonly used in recommendation systems, as input instead of the item description. Additionally, we apply improved deep learning model structures such as Self-Attention, Multi-head Attention, and Multi-Conv1D. We conducted experiments on various datasets with input and model structure changes. The results showed that when only the input was changed, MAE increased slightly compared to ADLFM due to accompanying information loss, resulting in decreased recommendation performance. However, the average learning speed per epoch significantly improved as the amount of information to be processed decreased. When both the input and the model structure were changed, the best-performing Multi-Conv1d structure showed similar performance to ADLFM, sufficiently counteracting the information loss caused by the input change. We conclude that G-ADLFRM is a new, lightweight, and generalizable model that maintains the performance of the existing ADLFM while enabling fast learning and inference.

• Journal of the Korea Institute of Building Construction
• /
• v.19 no.1
• /
• pp.9-18
• /
• 2019

Concrete, which is the most widely used building material in modern times, has been improved not only in strength but also in structural performance such as increase in toughness and ductility, weight reduction, and improvement in quality of human life. Due to the surge in demand for the building, there is a tendency to be used variously from architectural panel and architecture to interior accessories. In Korea, a light-transmitting concrete, LEFC(Light Emotion Friendly Concrete), that insert plastic rods to stimulate emotional sensation through the combination of light and concrete has developed. In previous research, it was confirmed that the use of a synthetic foam agent rather than an animal foam agent did not cause a fogging phenomenon. In this study, lightweight by applying foaming agent to LEFC and two types of fiber (Nylon Fiber, Polyvinyl Alcohol) were compared to achieve to investigate the fiber to be applied in future. An equation that can predict the loss and adhesion reduction of the concrete section according to the diameter of the rod (5mm, 10mm) and the interval (10mm, 15mm, 20mm) was proposed.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.89-95
• /
• 2024

Nowadays, the competitiveness of a company depends on the ability of all organizational members to share and utilize the organizational knowledge accumulated by the organization. As if to prove this, the world is now focusing on ChetGPT service using generative AI technology based on LLM (Large Language Model). However, it is still difficult to apply the ChetGPT service to work because there are many hallucinogenic problems. To solve this problem, sLLM (Lightweight Large Language Model) technology is being proposed as an alternative. In order to construct sLLM, corporate data is essential. Corporate data is the organization's ERP data and the company's office document knowledge data preserved by the organization. ERP Data can be used by directly connecting to sLLM, but office documents are stored in file format and must be converted to data format to be used by connecting to sLLM. In addition, there are too many technical limitations to utilize office documents stored in file format as organizational knowledge information. This study proposes a method of storing office documents in DB format rather than file format, allowing companies to utilize already accumulated office documents as an organizational knowledge system, and providing office documents in data form to the company's SLLM. We aim to contribute to improving corporate competitiveness by combining AI technology.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_3
• /
• pp.997-1008
• /
• 2023

Ship detection at sea can be performed in various ways. In particular, satellites can provide wide-area surveillance, and Synthetic Aperture Radar (SAR) imagery can be utilized day and night and in all weather conditions. To propose an efficient ship detection method from SAR images, this study aimed to apply the You Only Look Once Version 5 (YOLOv5) model to Sentinel-1 images and to analyze the difference between individual vs. integrated models and the accuracy characteristics by polarization. YOLOv5s, which has fewer and lighter parameters, and YOLOv5x, which has more parameters but higher accuracy, were used for the performance tests (1) by dividing each polarization into HH, HV, VH, and VV, and (2) by using images from all polarizations. All four experiments showed very similar and high accuracy of 0.977 ≤ AP@0.5 ≤ 0.998. This result suggests that the polarization integration model using lightweight YOLO models can be the most effective in terms of real-time system deployment. 19,582 images were used in this experiment. However, if other SAR images,such as Capella and ICEYE, are included in addition to Sentinel-1 images, a more flexible and accurate model for ship detection can be built.