Kisti Institutional Repository
Not a member yet
7415 research outputs found
Sort by
A Dual Control Approach for Indirect Configuration Propagation with Energy-Efficient Scheduling in Multi-agent Networking Systems
No full textThis paper presents a dual control approach for indirect system configuration propagation with energy-efficient agent scheduling. The proposed method influences the MNS (Multi-agent Networking Systems) operation by indirectly propagating the system configuration within the framework of local rules. Also, the proposed method adapts agent’s operational state according to the convergence rate of configuration propagation in order to balance energy consumption among agents in the MNS. Finally, we propose an optimal timing control for sequent input. Using the operation state control model, the gateway agent determines the optimal timing to give next input based on the value of the operation state. Simulation results are performed to demonstrate the superiority of the proposed method and we observe that the proposed scheme is less susceptible to error and shows more robust performance than the consensus method in an error-prone environment
CycleGAN을 이용한 야간 상황 물체 검출 알고리즘
No full textRecently, image-based object detection has made great progress with the introduction of Convolutional Neural Network (CNN). Many trials such as Region-based CNN, Fast R-CNN, and Faster R-CNN, have been proposed for achieving better performance in object detection. YOLO has showed the best performance under consideration of both accuracy and computational complexity. However, these data-driven detection methods including YOLO have the fundamental problem is that they can not guarantee the good performance without a large number of training database. In this paper, we propose a data sampling method using CycleGAN to solve this problem, which can convert styles while retaining the characteristics of a given input image. We will generate the insufficient data samples for training more robust object detection without efforts of collecting more database. We make extensive experimental results using the day-time and night-time road images and we validate the proposed method can improve the object detection accuracy of the night-time without training night-time object databases, because we converts the day-time training images into the synthesized night-time images and we train the detection model with the real day-time images and the synthesized night-time images
First Measurement of the Hubble Constant from a Dark Standard Siren using the Dark Energy Survey Galaxies and the LIGO/Virgo Binary–Black-hole Merger GW170814
No full textWe present a multi-messenger measurement of the Hubble constant H0 using the binary-black-hole merger GW170814 as a standard siren, combined with a photometric redshift catalog from the Dark Energy Survey (DES). The luminosity distance is obtained from the gravitational wave signal detected by the Laser Interferometer Gravitational-Wave Observatory (LIGO)/Virgo Collaboration (LVC) on 2017 August 14, and the redshift information is provided by the DES Year 3 data. Black hole mergers such as GW170814 are expected to lack bright electromagnetic emission to uniquely identify their host galaxies and build an object-by-object Hubble diagram. However, they are suitable for a statistical measurement, provided that a galaxy catalog of adequate depth and redshift completion is available. Here we present the first Hubble parameter measurement using a black hole merger. Our analysis results in H0=75_{-32}^{+40}km s-1Mpc-1, which is consistent with both SN Ia and cosmic microwave background measurements of the Hubble constant. The quoted 68% credible region comprises 60% of the uniform prior range [20, 140] km s-1 Mpc-1, and it depends on the assumed prior range. If we take a broader prior of [10, 220] km s-1 Mpc-1, we find H0=78_{-24}^{+96} km s-1 Mpc-1 (57% of the prior range). Although a weak constraint on the Hubble constant from a single event is expected using the dark siren method, a multifold increase in the LVC event rate is anticipated in the coming years and combinations of many sirens will lead to improved constraints on H0
Study on Geant4 Simulation Toolkit Using a Low-Energy Physics Profiling System
No full textThe Geant4 simulation toolkit needs to be optimized for central processing unit (CPU) time and memory size in order to be used for a high luminosity Large Hadron Collider (LHC) experiment. We have developed a low-energy physics profiling system and studied Geant4 with respect to both software and hardware. The results of profiling with Geant4 for low-energy physics were provided to the users and developers of the Geant4 simulation toolkit. In the case of the software, we performed profiling for dependence on different Geant4 versions and provided feedback to Geant4 developers. Profiling for dependence on the number of events showed their linear dependency. In the case of the hardware, we performed profiling for machine dependency on the KISTI-4 supercomputer. This action showed that machines were homogenous. The results obtained in this study will serve as a reference for Geant4 users
Development of ASEAN Network Model on Information Literacy
Get PDFThis study aimed at overviewing the situation of information literacy education and research in the Association of Southeast Asian Nations (ASEAN) region, and developing an ASEAN network model on information literacy. This research used documentary and qualitative research methods. Key resources consisted of twenty bibliometric studies and related documents and two groups of key persons. The first group consisted of twenty-seven purposive key persons from eight countries, and the second group consisted of seven key persons from five countries. The research instruments comprised a data collection form and focus group/ interviewing forms. Data was collected by focus group discussion and online interviews, and qualitative content analysis was used in data analysis and presented descriptively. Research findings showed that: 1) information literacy education and research in the ASEAN region varied across countries and placed importance on the educational context. Singapore was found to be the most leading and productive country in ASEAN in information literacy with the highest number of journal articles on the international scale, and was among the most contributing groups at the regional and global level; 2) the ASEAN Network on Information Literacy (ASEAN-NIL) has been developed as a model with its principles, objectives, management system, activities, and promotion strategies. Its strengths are an integrated scope, multidimensional orientation, and interdisciplinary and collaborative partnerships at the national, regional, and international level, suitable for the ASEAN context, the online environment, and the digital educational ecosystem
ASTI MARKET INSIGHT 76: 수상 태양광
No full text1. 수상태양광은 산림자원 감소와 생태계 및 경관 훼손 등의 부정적 영향을 미치는 육상태양광과 달리 상대적으로 환경영향이 적어 빠른 사업 추진이 가능하며, 발전효율이 육상 태양광 대비 10 % 가량 높음에 따라 최근 유휴수면을 대산으로 각광받고 있는 기술이다.
2. 국내에서는 2009년을 기점으로 국내에서 수상태양광 발전에 대한 관심이 증폭되면서 다양한 형태의 구조물이 등장하였으며, 한국농어촌공사와 한국수자원공사의 담수자원을 활용한 수상태양광이 많이 건설되고 있다.
3. 국내 한국농어촌공사에서 발주를 진행하는 저수지 영역, 한국수자원공사에서 진행하는 댐 수면, 정부에서 운영하는 해수면 영역에 대한 잠재량을 기준으로 시장을 추론하여 보면 연평균 성장률은 24 %로 2025년 1조8천억 원의 규모를 형성할 것으로 추산된다.
4. 하지만 해수면을 이용한 수상태양광의 경우 기존 담수시설용 실증기준을 적용하기에는 많은 환경변수가 있을 것으로 판단되며, 부가적으로 안전성, 내구성, 경제성, 환경성이 확보된 기술의 적용이 필요할 것이다
ASTI MARKET INSIGHT 88: 위치정보 관련 시장 동향 및 전망
No full text1. 위치정보란 이동성이 있는 물건 또는 개인이 특정한 시간에 존재하거나 존재하였던 장소에 관한 정보로서 전기통신설비 및 전기통신회선 설비를 이용하여 수집된 것으로 정의할 수 있다.
2. 2018년 개정된 위치정보법에서는 사물위치정보사업에 대해 신고제를 적용하고, 이동성 있는 물건의 위치정보 수집·이용·제공시 사전 동의를 받지 않아도 되며, 소상공인 및 1인 창조기업의 경우 사전신고가 면제된다.
3. 한국인터넷진흥원에 따르면 국내 위치정보 관련 시장 규모는 2020년 2조 2,827억 원에서 2021년 2조 6,279억 원으로 성장하였으며, 2022년에는 3조 550억 원으로 확대될 것으로 전망된다.
4. Marketsandamarkets에 따르면 세계 위치정보 관련 시장 규모는 2020년 61억 달러에서 2025년에는 169억 달러로 연평균 22.6% 성장할 것으로 전망된다.
5. 4차 산업 시대가 도래하면서 사물인터넷(IoT), 빅데이터, 인공지능(AI), 자율주행, 로봇자동화 등 첨단 산업에서 위치정보의 활용이 미래 성장 동력으로 대두되고 있으나 관련 인프라 구축에 많은 자원이 소모되므로 민간·공공의 축적된 측위 DB를 개방하여 기술기반의 중소·스타트업이 관련 산업에 진출할 수 있는 기반을 제공해야 한다고 판단된다
Knowledge Model for Disaster Dataset Navigation
Get PDFIn a situation where there are multiple diverse datasets, it is essential to have an efficient method to provide users with the datasets they require. To address this suggestion, necessary datasets should be selected on the basis of the relationships between the datasets. In particular, in order to discover the necessary datasets for disaster resolution, we need to consider the disaster resolution stage. In this paper, in order to provide the necessary datasets for each stage of disaster resolution, we constructed a disaster type and disaster management process ontology and designed a method to determine the necessary datasets for each disaster type and disaster management process step. In addition, we introduce a method to determine relationships between datasets necessary for disaster response. We propose a method for discovering datasets based on minimal relationships such as "isA," "sameAs," and "subclassOf." To discover suitable datasets, we designed a knowledge exploration model and collected 651 disaster-related datasets for improving our method. These datasets were categorized by disaster type from the perspective of disaster management. Categorizing actual datasets into disaster types and disaster management types allows a single dataset to be classified as multiple types in both categories. We built a knowledge exploration model on the basis of disaster examples to ensure the configuration of our model
현실 환경에서 증강현실과 딥러닝을 활용한 M&S 모델 증강가시화
No full textThis paper proposes a new method to effectively visualize modeling & simulation (M&S) results in a real environment using augmented reality (AR) and deep learning. The proposed approach makes it possible to dynamically generate an M&S analysis space of the real environment, to recognize real objects by using a deep learning technique, and to place the analyzed M&S results onto them. In order to construct an M&S space dynamically, we perform area learning on the real space using a smart device supporting RGB-D camera. In addition, real objects are recognized through deep learning-based object detection. Spatial mapping and user interaction are conducted to match the recognized real object with corresponding M&S model in the mobile AR environment. A proof-of-concept system was developed to show the advantage and feasibility of the proposed method. Therefore, the proposed approach can be used for seamlessly integrating M&S models into various real spaces and for reviewing M&S results more consistently and effectively