• The Transactions of the Korea Information Processing Society
• /
• v.13 no.3
• /
• pp.111-121
• /
• 2024

Recently, thanks to the development of artificial intelligence technologies such as image recognition, research on unmanned aerial vehicles is being actively conducted. In particular, related research is increasing in the field of military drones, which costs a lot to foster professional pilot personnel, and one of them is the study of an intelligent framework for autonomous mission performance of reconnaissance drones. In this study, we tried to design an intelligent framework for unmanned aerial vehicles using the methodology of designing an intelligent framework for service robots. For the autonomous mission performance of unmanned aerial vehicles, the intelligent framework and unmanned aerial vehicle module must be smoothly linked. However, it was difficult to provide interworking for drones using periodic message protocols with model-based interfaces of intelligent frameworks for existing service robots. First, the message model lacked expressive power for periodic message protocols, followed by the problem that interoperability of asynchronous data exchange methods of periodic message protocols and intelligent frameworks was not provided. To solve this problem, this paper proposes a message model extension method for message periodic description to secure the model's expressive power for the periodic message model, and proposes periodic and asynchronous data exchange methods using the extended model to provide interoperability of different data exchange methods.

• Journal of Korea Society of Industrial Information Systems
• /
• v.21 no.4
• /
• pp.67-78
• /
• 2016

Recently many countries around the world are actively promoting smart city projects to address various urban problems such as traffic congestion, housing shortage, and energy scarcity. Due to development of the Internet of Things (IoT), the development of a smart city with sustainability, convenience, and environment-friendliness was enabled through the effective control and reuse of urban resources. The purpose of this study is to analyze the technical trends of IoT and present a development plan for smart city which is one of the applications of the IoT. To this end, the news articles of the Electronic Times between 2013 and 2015were analyzed using the text mining technique and smart city development cases of other countries were investigated. The analysis results revealed the close relationships of big data, cloud, platforms, and sensors with smart city. For the successful development of a smart city, first, all the interested parties in the city must work together to create new values throughout the entire process of value chain. Second, they must utilize big data and disclose public data more actively than they are doing now. This study has made academic contribution in that it has presented a big data analysis method and stimulated follow-up studies. For the practical contribution, the results of this study provided useful data for the policy making of local governments and administrative agencies for smart city development. This study may have limitations in the incorporation of the total trends because only the news articles of the Electronic Times were selected to analyze the technical trends of the IoT.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1015-1023
• /
• 2022

The Entry-Descent-Landing process of a lander involves many environmental and technical challenges. To solve these problems, recently, terrestrial relative navigation (TRN) technology has been essential for landers. TRN is a technology for estimating the position and attitude of a lander by comparing Inertial Measurement Unit (IMU) data and image data collected from a descending lander with pre-built reference data. In this paper, we present a method for generating descent dataset and extracting landmarks, which are key elements for developing TRN technologies to be used on Mars. The proposed method generates IMU data of a descending lander using a simulated Mars landing trajectory and generates descent images from high-resolution ortho-map and digital elevation map through a ray tracing technique. Landmark extraction is performed by an area-based extraction method due to the low-textured surfaces on Mars. In addition, search area reduction is carried out to improve matching accuracy and speed. The performance evaluation result for the descent dataset generation method showed that the proposed method can generate images that satisfy the imaging geometry. The performance evaluation result for the landmark extraction method showed that the proposed method ensures several meters of positioning accuracy while ensuring processing speed as fast as the feature-based methods.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.137-142
• /
• 2024

Recently, deep learning technology has become those methods as de facto standards in the area of medical data representation. But, deep learning inherently requires a large amount of training data, which poses a challenge for its direct application in the medical field where acquiring large-scale data is not straightforward. Additionally, brain signal modalities also suffer from these problems owing to the high variability. Research has focused on designing deep neural network structures capable of effectively extracting spectro-spatio-temporal characteristics of brain signals, or employing self-supervised learning methods to pre-learn the neurophysiological features of brain signals. This paper analyzes methodologies used to handle small-scale data in emerging fields such as brain-computer interfaces and brain signal-based state prediction, presenting future directions for these technologies. At first, this paper examines deep neural network structures for representing brain signals, then analyzes self-supervised learning methodologies aimed at efficiently learning the characteristics of brain signals. Finally, the paper discusses key insights and future directions for deep learning-based brain signal analysis.

• Journal of Aerospace System Engineering
• /
• v.14 no.5
• /
• pp.83-90
• /
• 2020

Disasters are causing significant damage to the lives and property of our society and are recognized as social problems that need to be solved nationally and globally. The 4th industrial revolution technologies affecting society as a whole such as the Internet of Things(IoT), Artificial Intelligence(AI), Drones(Unmanned Aerial Vehicles), and Big Data are continuously absorbed into the disaster and safety industries as scientific and technological tools for solving social problems. Very soon, twenty-nine domestic UAV-related organizations/companies will complete the construction of a multicopter type small UAV integrated system ('17~'20) that can be operated at disaster and security sites. The current work considers and proposes the mid-to-long term R&D direction of disaster UAV as a strategic asset of the national disaster response system. First, the trends of disaster and safety industry and policy are analyzed. Subsequently, the development status and future plans of small UAV, securing shortage technology, and strengthening competitiveness are analyzed. Finally, step-by-step R&D direction of disaster UAV in terms of development strategy, specialized mission, platform, communication, and control and operation is proposed.

• Journal of KIISE:Databases
• /
• v.37 no.5
• /
• pp.238-249
• /
• 2010

Flash memory has been studied as a storage medium in order to improve the performance of the system using its high computing speed in the DBMS field where frequent data access is needed. The most difficulty using the flash memory is the performance degradation and the life span shortening of flash memory coming from inefficient in-place update. Log based approaches have been studied to solve inefficient in-place update problem in the DBMS where write operations occur in smaller size of data than page frequently. However the existing log based approaches suffer from the frequent merging operations, which are the principal cause of performance deterioration. Thus is because their fixed log area management can not guarantee a sufficient space for logs. In this paper, we propose non-fixed log area management technique that can minimize the occurrence of the merging operations by promising an enough space for logs. We also suggest the cost calculation model of the optimal log sector number minimizing the system operation cost in a block. In experiment, we show that our non-fixed log area management technique can have the improved performance compared to existing approaches.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.3
• /
• pp.48-55
• /
• 2015

This paper studies how to solve a problem caused by M2M terminals sending a few data based on $3^{rd}$ Generation Partnership Project(3GPP) Long Term Evolution-Advanced(LTE-A) system and then it is analyzed, proposed, and introduced into the techniques. Especially, it is introduced solution for the lack of Random Access Channel and an increasing number of latency caused by countless M2M devices. It is proposed the technology for M2M grouping as well as allowable access probability according to user type. As it decreases the number of terminal by grouping M2M devices to try random access at PRACH, it can be reduced collision between Cellular users and M2M devices. So, it is proved that the proposed mechanism can solve the increasing average latency of random access on system coexisting Cellular users and M2M devices through simulations.

• KIISE Transactions on Computing Practices
• /
• v.21 no.3
• /
• pp.269-274
• /
• 2015

As smart device companies such as Google or Apple develop competitive mobile-based automotive operating systems and navigation systems, the range of choice for users in such markets is expanding. Navigation systems equipped with mobile operating systems have increased convenience for users. However, since an API for the POI databases used in navigation systems doesn't exist, the number of applications using POI data is insufficient. In this paper, we designed and implemented system calls for navigation operating systems with a focus on POI search, in order to resolve such limitations. The system calls support set-based POI search functions, and therefore provides solutions to search performance degradation problems caused by false inputs. As a result of performance evaluation, not only did the search performance improve, but there was also no problem in applying APIs in applications.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.39 no.5
• /
• pp.391-395
• /
• 2015

A 15-Lead ECG has been developed to diagnose posterior wall diseases of the heart that a 12-Lead ECG cannot diagnose. However, 15-Lead ECG data for developing heart-diseases-detecting algorithm are limited, and previous ECG simulators cannot predict the ECG waveform according to the changes in electrode. To solve these problems, the lumped parameter model (LPM), which divides the heart into 15 sections with varying electrical capacitance and electrical resistance. To imitate the electrical conduction in the heart, each node was connected to a current source and delivered the specific current considering the positions and time delay. The purpose of this study is to acquire the waveform that can be used in an ECG by delivering the specific current to LPM.

• Journal of Digital Convergence
• /
• v.19 no.6
• /
• pp.251-258
• /
• 2021

In this paper, we proposed an encoder-decoder model utilizing residual learning to improve the accuracy of the U-Net-based semantic segmentation method. U-Net is a deep learning-based semantic segmentation method and is mainly used in applications such as autonomous vehicles and medical image analysis. The conventional U-Net occurs loss in feature compression process due to the shallow structure of the encoder. The loss of features causes a lack of context information necessary for classifying objects and has a problem of reducing segmentation accuracy. To improve this, The proposed method efficiently extracted context information through an encoder using residual learning, which is effective in preventing feature loss and gradient vanishing problems in the conventional U-Net. Furthermore, we reduced down-sampling operations in the encoder to reduce the loss of spatial information included in the feature maps. The proposed method showed an improved segmentation result of about 12% compared to the conventional U-Net in the Cityscapes dataset experiment.