연구성과로 돌아가기
2023 연구성과 (267 / 285)
※ 컨트롤 + 클릭으로 열별 다중 정렬 가능합니다.
Excel 다운로드
| WoS | SCOPUS | Document Type | Document Title | Abstract | Authors | Affiliation | ResearcherID (WoS) | AuthorsID (SCOPUS) | Author Email(s) | Journal Name | JCR Abbreviation | ISSN | eISSN | Volume | Issue | WoS Edition | WoS Category | JCR Year | IF | JCR (%) | FWCI | FWCI Update Date | WoS Citation | SCOPUS Citation | Keywords (WoS) | KeywordsPlus (WoS) | Keywords (SCOPUS) | KeywordsPlus (SCOPUS) | Language | Publication Stage | Publication Year | Publication Date | DOI | JCR Link | DOI Link | WOS Link | SCOPUS Link |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ○ | Conference paper | Lg= 60 nm In0.53Ga0.47As MBCFETs: From gm-max= 13.7 mS/μm and Q = 180 to virtual-source modeling | In this paper, we report scalable 5 -level stacked gate-all-around (GAA) In 0.53 Ga0.47As multi-bridge channel FETs (MBCFETs), with careful attention paid to fluorine migration. At its heart, we maintained temperature of all the unit process steps below 300 °C and inserted an n-InP ledge into a top In0.52Al0.48As sacrificial layer to suppress F- -induced donor passivation. In addition, we used a selectively regrown n+In0.53Ga0.47 As contact formation by MOCVD and precision dry etching. The dry etching process resulted in a highly vertical etching slope along both the S/D and Wg directions. The fabricated Lg=60nm MBCFET showed a record combination of S=76mV/dec, gm-max=13.7 mS/mum, ION=2.24mA/mum and Q=180 at VDS=0.5V. © 2023 JSAP. | Yoo, J.-H.; Jo, H.-B.; Lee, I.-G.; Choi, S.-M.; Baek, J.-M.; Lee, S.T.; Jang, H.; Kong, M.W.; Kim, H.H.; Lee, H.J.; Kim, H.-J.; Jeong, H.-S.; Park, W.-S.; Ko, D.H.; Shin, S.H.; Kwon, H.-M.; Kim, S.K.; Kim, J.G.; Yun, J.; Kim, T.; Shin, K.-Y.; Kim, T.-W.; Shin, J.-K.; Lee, J.-H.; Shin, C.-S.; Seo, K.-S.; Kim, D.-H. | Kyungpook National University, South Korea; Kyungpook National University, South Korea; Kyungpook National University, South Korea; Kyungpook National University, South Korea; Kyungpook National University, South Korea; Korea Advanced Nano Fab Center, South Korea; Korea Advanced Nano Fab Center, South Korea; Korea Advanced Nano Fab Center, South Korea; Kyungpook National University, South Korea; Kyungpook National University, South Korea; Kyungpook National University, South Korea; Kyungpook National University, South Korea; Kyungpook National University, South Korea; Yonsei University, South Korea; Korea Polytechnics, South Korea; Korea Polytechnics, South Korea; Qsi; Qsi; Qsi; Qsi; University of Ulsan, South Korea; University of Ulsan, South Korea; Kyungpook National University, South Korea; Kyungpook National University, South Korea; Korea Advanced Nano Fab Center, South Korea; Korea Advanced Nano Fab Center, South Korea; Kyungpook National University, South Korea | 57545572700; 57202871742; 37016357200; 57825819100; 57189694750; 57221766125; 55860474200; 59109130700; 58530978900; 59106441600; 57202516002; 57200366272; 57222957219; 16240618000; 57217467919; 55549386600; 57221604960; 59087926000; 57221602607; 59603995600; 57221120602; 57203495132; 7402723873; 55690077600; 57201540732; 58711653100; 57212363794 | Digest of Technical Papers - Symposium on VLSI Technology | 0743-1562 | 2023-June | 0.61 | 2025-06-25 | 1 | Dry etching; Indium phosphide; Semiconducting indium phosphide; Semiconductor alloys; 5-level; Contact formation; Dry etching process; Gate-all-around; Process steps; Sacrificial layer; Source models; Stacked gate; Vertical etching; Virtual sources; III-V semiconductors | English | Final | 2023 | 10.23919/vlsitechnologyandcir57934.2023.10185250 | 바로가기 | 바로가기 | ||||||||||||||||||
| ○ | Conference paper | Lightweighted FPGA Implementation of Even-Odd-Buffered Active Noise Canceller with On-Chip Convolution Acceleration Units | To make the acoustic signal that was processed by the noise canceller sound more natural for users [1]-[3], the delay in anti-noise generation should be reduced. For a single buffer, processing delay occurrs because it is impossible to write input signals while the processor is processing the data. when interfering with anti-noise and output signal, this processing delay creates additional buffering overhead to match the phase. The processing delay can be minimized using an Even-/Odd-buffer Structure to alternately read and write operations. In addition, the differences between the two methods of noise cancellation (FFT-based noise cancellation and adaptive algorithm) are compared in terms of output signal quality, processing time, and power consumption. As a result, using an Even-/Odd-buffer, reduced the processing delay of a single buffer. The FFT-based noise canceling method experienced fewer errors than the adaptive noise canceling method. © 2023 IEEE. | Park, Seunghyun; Park, Daejin | Kyungpook National University, School of Electronic and Electrical Engineering, South Korea; Kyungpook National University, School of Electronic and Electrical Engineering, South Korea | 57903951400; 55463943600 | boltanut@knu.ac.kr; | 2023 International Conference on Electronics, Information, and Communication, ICEIC 2023 | 1.58 | 2025-06-25 | 3 | Adaptive algorithm; Even-Odd-buffer; FFT; Noise cancelling | Acoustic noise; Data handling; Fast Fourier transforms; Field programmable gate arrays (FPGA); Spurious signal noise; Active noise; Antinoise; Even-odd-buffer; FPGA implementations; FPGAs implementation; Noise cancellation; Noise canceller; Noise cancelling; Processing delay; Single buffers; Adaptive algorithms | English | Final | 2023 | 10.1109/iceic57457.2023.10049949 | 바로가기 | 바로가기 | ||||||||||||||||||
| ○ | Conference paper | Lightweighted Shallow CTS Techniques for Checking Clock Tree Synthesizable Paths and Optimizing Clock Tree in RTL Design Time | The microcontroller unit (MCU) are mainly used in low power devices, which use limited energy sources such as batteries, energy harvesting, and wireless communications. Therefore, reducing the operating power of the MCU is important to improve energy efficiency by extending battery life or minimizing energy consumption. The MCU is one of the chip designs composed of digital integrated circuits. The clock signal is important element to the MCU. The clock tree, which consists of the clock signal, is directly related to MCU low power operation and performance improvement. Also, chip verification process is important role to improve performance of overall system and reliability to the MCU. However, as the degree of integation of chips, the chip verification process increases complexity and time-comsumption to process many data. Currently, many users dependent on licensed electronic design automation (EDA) tools to ensure high accuracy, minimizing errors in circuit design and improving reliability. The use of licensed EDA tool puts a burden on users including high costs, limited license, difficulty in customization, slow speed, etc. An effective approach to avoid problems by using licensed EDA tools proceeds verification that is unrestricted license and customization for possible using only a register transfer level (RTL) source. In this paper, we propose to predict roughly pre-estimated CTS results using an RTL source in which temporary logic using random buffer insertion is placed before the route process. This paper contributes to reducing MCU operating power and hardware area by performing optimized CTS and minimizing resources according to the RTL structure to be designed. © 2023 IEEE. | Kwon, Nayoung; Park, Daejin | Kyungpook National University, School of Electronic and Electrical Engineering, South Korea; Kyungpook National University, School of Electronic and Electrical Engineering, South Korea | 57638935700; 55463943600 | boltanut@knu.ac.kr; | International Conference on ICT Convergence | 2162-1233 | 0 | 2025-06-25 | 0 | clock tree synthesis (CTS); licensed EDA tool; low power; micro controller unit (MCU); placement and route (P&R); register transfer level (RTL); shallow CTS; synthesizable | Clocks; Computer aided software engineering; Controllers; Electronic design automation; Energy harvesting; Energy utilization; Integrated circuit design; Integrated circuit manufacture; Microcontrollers; Clock tree synthesis; Electronic design automation tools; Licensed electronic design automation tool; Low Power; Micro controller unit; Micro controller units; Placement and route; Register transfer level; Register-transfer level; Shallow clock tree synthesis; Synthesizable; Energy efficiency | English | Final | 2023 | 10.1109/ictc58733.2023.10393374 | 바로가기 | 바로가기 | |||||||||||||||||
| ○ | Conference paper | Liquid-Crystalline Lens Array-Based 3D/2D Switchable Augmented Reality Display System Using One-Shot Learning Model | An advanced three-dimensional/two-dimensional switchable augmented-reality system is proposed. A camera captures the images of real objects, and it is displayed as it is or the elemental image array generated from a captured image and estimated depth information via a one-shot learning model, is reconstructed by a liquid-crystalline lens array. © 2023 The Author (s). | Erdenebat, Munkh-Uchral; Khuderchuluun, Anar; Darkhanbaatar, Nyamsuren; Nam, Oh-Seung; Kim, Joon Hyun; Imtiaz, Shariar Md.; Kwon, Ki-Chul; Gil, Sang-Keun; Kim, Hak-Rin; Kim, Nam | School of Information and Communication Engineering, Chungbuk National University, 1 Chungdae-ro, Seowon-gu. Cheongju, Chungbuk, 28644, South Korea; School of Information and Communication Engineering, Chungbuk National University, 1 Chungdae-ro, Seowon-gu. Cheongju, Chungbuk, 28644, South Korea; School of Information and Communication Engineering, Chungbuk National University, 1 Chungdae-ro, Seowon-gu. Cheongju, Chungbuk, 28644, South Korea; School of Information and Communication Engineering, Chungbuk National University, 1 Chungdae-ro, Seowon-gu. Cheongju, Chungbuk, 28644, South Korea; School of Information and Communication Engineering, Chungbuk National University, 1 Chungdae-ro, Seowon-gu. Cheongju, Chungbuk, 28644, South Korea; School of Information and Communication Engineering, Chungbuk National University, 1 Chungdae-ro, Seowon-gu. Cheongju, Chungbuk, 28644, South Korea; School of Information and Communication Engineering, Chungbuk National University, 1 Chungdae-ro, Seowon-gu. Cheongju, Chungbuk, 28644, South Korea; Department of Electronics Engineering, Suwon University, 17 Wauan-gil, Bongdam-eup, Gyeonggi, Hwaseong, 18323, South Korea; School of Electronics Engineering, Kyungpook National University, 80 Daehak-ro, Buk-gu, Daegu, 41566, South Korea; School of Information and Communication Engineering, Chungbuk National University, 1 Chungdae-ro, Seowon-gu. Cheongju, Chungbuk, 28644, South Korea | 36166588400; 57203638261; 57203635381; 57344811300; 57347716000; 57213601979; 7201503212; 55663924400; 7410124944; 35494120000 | namkim@chungbuk.ac.kr; | 3D Image Acquisition and Display: Technology, Perception and Applications in Proceedings Optica Imaging Congress, 3D, COSI, DH, FLatOptics, IS, pcAOP 2023 | 0 | 2025-06-25 | 0 | Learning systems; Liquid crystals; Three dimensional displays; Augmented reality systems; Crystalline lens; Display system; Learning models; Lens array; Liquid crystalline; One-shot learning; Real objects; Switchable; Two-dimensional; Augmented reality | English | Final | 2023 | 10.1364/3d.2023.dw4a.5 | 바로가기 | 바로가기 | |||||||||||||||||||
| ○ | Article | Local Region-Based Hand Segmentation Using Deep Learning | Segmenting objects through the Digital Image Processing makes it difficult to draw high accuracy in a complex background. In this paper, we propose a method of using deep-learning segmentation to segment objects in complex environments, providing ROI(Region of Interest) of images as a pre-processing, and Thresholding in the post-processing. The proposed method is to detect the position of the hand in a complex image using a deep learning-based object recognition algorithm employing the YOLOv4 model; to expand the ROI so that the deep learning-based segmentation techniques using the U-Net model can be applied locally, not on the entire image; and to process by Thresholding through Otsu's Binarization method. We applied the proposed algorithm to hand images with complex backgrounds and verified the effectiveness of the algorithm by measuring the IoU values of the masks of correct answers and results. © 2023, Korean Institute of Communications and Information Sciences. All rights reserved. | Jeon, Junhyun; Kim, Tae-Hun; Jeong, Yoosoo; Park, Kil-Houm | Department of Electronic and Electrical Engineering, Kyungpook National University, South Korea; DIPVISION, Poland; Electronics and Telecommunications Research Institute(ETRI), South Korea; Department of Electronic and Electrical Engineering, Kyungpook National University, South Korea | 58965765300; 55696523100; 57193450818; 35776805000 | khpark@ee.knu.ac.kr; | Journal of Korean Institute of Communications and Information Sciences | 1226-4717 | 48 | 9 | 0 | 2025-06-25 | 0 | Deep Learning; Hand; Object Detection; Object Segmentation | Korean | Final | 2023 | 10.7840/kics.2023.48.9.1135 | 바로가기 | 바로가기 | ||||||||||||||||
| ○ | Book chapter | Localization with Wi-Fi Ranging and Built-in Sensors: Self-Learning Techniques | Securing precise distance measurements from nearby reference nodes is a critical task in determining the performance of range-based positioning solutions. However, the multipath propagation characteristics of wireless channels render it difficult to obtain precise ranging results. In this context, this chapter utilizes machine learning techniques that extract useful features from Wi-Fi measurements to identify channel conditions and thus produce enhanced ranging results. Specifically, two neural networks (NN) are designed to perform ranging procedures for signal strength-based and round-trip time-based ranging scenarios. Furthermore, self-learning techniques that train the proposed NN-based ranging models with unlabeled training data are discussed in detail. The effectiveness of the proposed ranging models and self-learning techniques is extensively verified using a real-time positioning application. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023. | Choi, Jeongsik; Choi, Yang-Seok; Talwar, Shilpa | School of Electronics Engineering, Kyungpook National University, Daegu, South Korea; Intel Labs, Intel Corporation, Hillsboro, OR, United States; Intel Labs, Intel Corporation, Santa Clara, CA, United States | 58534394200; 37058389600; 57216999310 | jeongsik.choi@knu.ac.kr; | Machine Learning for Indoor Localization and Navigation | 0.7 | 2025-06-25 | 1 | Inertial sensors; Received signal strength; Round-trip time; Unsupervised learning; Wi-Fi ranging | English | Final | 2023 | 10.1007/978-3-031-26712-3_5 | 바로가기 | 바로가기 | |||||||||||||||||||
| ○ | Article | Locking Screw Insertion Path Guidance System in Minimally Invasive Fracture Reduction Surgery | In this paper, we propose a system to detect the location and direction of locking screw insertion using only a fluoroscopic image and guide screw insertion path, with a device that combines a SCARA robot arm and a conical remote center-of-motion (RCM) mechanism. In the proposed system, the target location and direction for screw insertion were derived from the fluoroscopic image of an intramedullary nail. The insertion guidance device comprise the SCARA robot arm and the RCM mechanism. The insertion guidance device calculated the current indicated path through forward kinematics analysis, and the target path and the current path were transmitted to the orthopedic surgeon through the control program. Moreover, the insertion guidance device set the insertion position and direction independently. The orthopedic surgeon first adjusted the SCARA robot arm to match the insertion position value to the target value and then sequentially adjusted the RCM mechanism so that the insertion direction matched the target value. After determining the insertion position and direction, the orthopedic surgeon inserted the surgical instrument through the path guided by the insertion guidance device and fastened the screw to the bone fragments and the intramedullary nail. When the proposed screw insertion guidance system was applied to fracture surgery, the locking screw was fastened with a minimum number fluoroscopic image, thus reducing surgery time and radiation exposure. The validity of the proposed method was confirmed through simulations and experiments using a fracture model. © ICROS 2023. | Choi, Youn-Ho; Lee, Suk-Joong; Oh, Chang-Wug; Bang, Hyun Hee; Lee, Hyun Woo; Park, Chul Woo | Institute of Medical Device and Robot, Kyungpook National University, South Korea; Department of Orthopaedic Surgery, Gyeongsang National University College of Medicine and Gyeongsang National University Changwon Hospital, South Korea; Department of Orthopaedic Surgery, School of Medicine, Kyungpook National University, Kyungpook National University Hospital, South Korea; Institute of Medical Device and Robot, Kyungpook National University, South Korea; Institute of Medical Device and Robot, Kyungpook National University, South Korea; Institute of Medical Device and Robot, Kyungpook National University, South Korea | 58249710900; 57078230800; 22135834200; 57216785452; 59400287600; 23995581000 | pcw.mdrip@gmail.com; | Journal of Institute of Control, Robotics and Systems | 1976-5622 | 29 | 5 | 0 | 2025-06-25 | 0 | C-arm fluoroscopy; intramedullary nailing; locking screw guide; minimal invasive fracture surgery | Bone; Locks (fasteners); Remote control; Robotic arms; Screws; Surgery; Surgical equipment; C-arm fluoroscopy; Fluoroscopic images; Intramedullary nailing; Locking screw guide; Locking screws; Minimal invasive; Minimal invasive fracture surgery; Robot arms; SCARA robot; Screw insertions; Fracture | Korean | Final | 2023 | 10.5302/j.icros.2023.23.0014 | 바로가기 | 바로가기 | |||||||||||||||
| ○ | Book chapter | Locust attacks on crop plants and control strategies to minimize the extent of the problem | Locusts' invasion has been a major threat to crop plants and overall food security across the globe. Crop damage by locusts is even more threatening and devastating due to global pandemics such as COVID-19. Recently, the locust's attack has caused major damage to the agricultural and food crops and the value of this damage is estimated to cross billions of dollars. Although these locusts and grasshoppers are major feeders of crop plants, at the same time their populations are also important to maintain ecosystem functioning and balance. However, their overpopulation in certain seasons can cause major destruction of growing crops. Therefore, it is significantly important to check their populations, feeding activities, habitats, and crop damage in various regions so that effective management measures and strategies can be devised as per requirements and extent of the problem. The control and management measures mentioned in this chapter can greatly help the farmers and researchers to minimize the locust's invasion to achieve food security on a sustainable basis. © 2024 Apple Academic Press, Inc. | Athar, Tabinda; Fatima, Hina; Waris, Aisha A.; Aqsa; Kanwal, Nafisa; Majid, Farah | Institute of Soil and Environmental Sciences, University of Agriculture Faisalabad, Faisalabad, Pakistan; School of Applied Biosciences, Kyungpook National University, Daegu, South Korea; Institute of Soil and Environmental Sciences, University of Agriculture Faisalabad, Faisalabad, Pakistan; Department of Soil Science, University of Agriculture Faisalabad, Sub Campus Burewala, Vehari, Pakistan; Department of Soil Science, University of Agriculture Faisalabad, Sub Campus Burewala, Vehari, Pakistan; Department of Entomology, University of Agriculture, Faisalabad, Faisalabad, Pakistan | 57215817227; 59800323300; 57204141168; 58342359900; 58510900500; 58511597700 | Locust Outbreaks: Management and the World Economy | 1.92 | 2025-06-25 | 2 | Climate change; Cotton; Desert locust; Management tools; Red locust; Sustainability | English | Final | 2023 | 바로가기 | ||||||||||||||||||||||
| ○ | ○ | Article | Low-Power FPGA Realization of Lightweight Active Noise Cancellation with CNN Noise Classification | Active noise cancellation (ANC) is the most important function in an audio device because it removes unwanted ambient noise. As many audio devices are increasingly equipped with digital signal processing (DSP) circuits, the need for low-power and high-performance processors has arisen because of hardware resource restrictions. Low-power design is essential because wireless audio devices have limited batteries. Noise cancellers process the noise in real time, but they have a short secondary path delay in conventional least mean square (LMS) algorithms, which makes implementing high-quality ANC difficult. To solve these problems, we propose a fixed-filter noise cancelling system with a convolutional neural network (CNN) classification algorithm to accommodate short secondary path delay and reduce the noise ratio. The signal-to-noise ratio (SNR) improved by 2.3 dB with CNN noise cancellation compared to the adaptive LMS algorithm. A frequency-domain noise classification and coefficient selection algorithm is introduced to cancel the noise for time-varying systems. Additionally, our proposed ANC architecture includes an even-odd buffer that efficiently computes the fast Fourier transform (FFT) and overlap-save (OLS) convolution. The simulation results demonstrate that the proposed even-odd buffer reduces processing time by 20.3% and dynamic power consumption by 53% compared to the single buffer. | Park, Seunghyun; Park, Daejin | Kyungpook Natl Univ, Sch Elect & Elect Engn, Daegu 41566, South Korea | 57903951400; 55463943600 | ijjh0435@knu.ac.kr;boltanut@knu.ac.kr; | ELECTRONICS | 2079-9292 | 12 | 11 | 0.47 | 2025-06-25 | 3 | 5 | active noise cancellation; convolutional neural networks; even-odd buffer; low-power system design; digital signal processing | active noise cancellation; convolutional neural networks; digital signal processing; even–odd buffer; low-power system design | English | 2023 | 2023-06-02 | 10.3390/electronics12112511 | 바로가기 | 바로가기 | 바로가기 | ||||||||||||
| ○ | Conference paper | Low-to-mid spatial frequency wavefront error control for off-axis freeform three-mirror KASI-Deep Rolling Imaging Fast Telescope | We have developed the KASI-Deep Rolling Imaging Fast Telescope (K-DRIFT), adopting a 300 mm aperture off-axis freeform three-mirror design to detect faint and diffuse low-surface-brightness structures. By conducting the on-sky test observations and performing a series of simulations to analyze the performance of the K-DRIFT, we confirmed three main error sources causing optical performance degradation. The imaging performance of the K-DRIFT has successfully improved by correcting low-to-mid spatial frequency wavefront errors based on performance analysis results. This paper presents the K-DRIFT’s optical performance analysis algorithm and the optical performance improvement. © 2023 SPIE. | Lee, Gayoung; Kim, Yunjong; Kim, Daewook; Byun, Woowon; Choi, Changsu; Lee, Yongseok; Kim, Dohoon; Chang, Seunghyuk; Ko, Jongwan | Department of Astronomy, Kyungpook National University, Daegu, 41566, South Korea, Korea Astronomy and Space Science Institute, Daejeon, 34055, South Korea; Korea Astronomy and Space Science Institute, Daejeon, 34055, South Korea; Wyant College of Optical Sciences, University of Arizona, Tucson, 85721, AZ, United States; Korea Astronomy and Space Science Institute, Daejeon, 34055, South Korea; Korea Astronomy and Space Science Institute, Daejeon, 34055, South Korea; Korea Astronomy and Space Science Institute, Daejeon, 34055, South Korea, School of Space Research, Kyung Hee University, Yongin, 17104, South Korea; Green Optics Co., Ltd., Cheongju, 28126, South Korea; Center for Integrated Smart Sensors, Daejeon, 34141, South Korea; Korea Astronomy and Space Science Institute, Daejeon, 34055, South Korea, University of Science and Technology, Daejeon, 34055, South Korea | 57846551600; 57203310827; 57219213285; 57216758314; 18433745100; 36068040300; 57295662700; 12759772300; 18434193800 | gylee9097@kasi.re.kr; | Proceedings of SPIE - The International Society for Optical Engineering | 0277-786X | 12677 | 0 | 2025-06-25 | 0 | freeform; K-DRIFT; mid-spatial frequency error; misalignment; off-axis; optical performance analysis; telescope; three-mirror system | Mirrors; Telescopes; Wavefronts; Deep rolling; Freeforms; Frequency errors; KASI-deep rolling imaging fast telescope; Mid-spatial frequency error; Mirror systems; Misalignment; Off-axis; Optical performance analysis; Spatial frequency; Three-mirror system; Errors | English | Final | 2023 | 10.1117/12.2676309 | 바로가기 | 바로가기 | ||||||||||||||||
| ○ | Article | LSTM-Based Imitation Learning of Robot Manipulator Using Impedance Control | This paper proposes an imitation learning method based on long short-term memory (LSTM) to demonstrate robot manipulators using impedance control. An impedance controller controls the force and position of the robot manipulator. In this study, direct demonstrated position and force data for imitation learning of the robot were designed to be the reference input of the impedance controller. LSTM-based imitation learning methods enabled the robot to function as intended, even when its initial position was changed or other contact forces were applied according to the environment. The proposed method was verified by applying the writing task of the actual industrial robot manipulator that functions as the expert’s intention. © ICROS 2023. | Park, Sejun; Jo, Seonghyeon; Lee, Sangmoon | Department of Electronic and Electrical Engineering, Kyungpook National University, South Korea; Department of Electronic and Electrical Engineering, Kyungpook National University, South Korea; Department of Electronic and Electrical Engineering, Kyungpook National University, South Korea | 58097901300; 57220154785; 59510733500 | moony@knu.ac.kr; | Journal of Institute of Control, Robotics and Systems | 1976-5622 | 29 | 2 | 0.14 | 2025-06-25 | 1 | Character Writing Task; Imitation Learning; Impedance Control; LSTM; Robot Manipulator | Controllers; Flexible manipulators; Industrial robots; Learning systems; Modular robots; Robot applications; Character writing task; Contact forces; Force data; Imitation learning; Impedance control; Impedance controllers; Learning methods; Position data; Reference inputs; Robots manipulators; Long short-term memory | Korean | Final | 2023 | 10.5302/j.icros.2023.22.0218 | 바로가기 | 바로가기 | |||||||||||||||
| ○ | ○ | Proceedings Paper | Lunar Landing Site Selection using Machine Learning | The value of a planetary landing mission is critically dependent on the choice of landing sites which are generally influenced by several scientific and engineering constraints. The conventional method for Landing Site (LS) selection requires the manual analyses of individual sites involving a large talent force of multi-domain experts hence rendering the process cumbersome and expensive. Simultaneously, the currently employed methodology of recycling previously selected LSs for future missions would not be realized in the case of hitherto unexplored planets. As a consequence, machine learning algorithms are being actively explored in aiding effective space exploration. However, their application to selection of LSs that satisfy the scientific and engineering requirements of a mission have not yet been explored. In this paper, we propose an end-to-end Landing Site selection methodology using Moon as a case study and employing Hierarchical clustering of regions based on their altitude, expandable to other planets. Furthermore, we enforce commonly used constraints on the potential sites and select final sites for landing based on the user provided scientific and engineering constraints. With this approach, the LS selection process is simplified and the temporal requirement is reduced. | Darlan, Daison; Ajani, Oladayo S.; Mallipeddi, Rammohan | Kyungpook Natl Univ, Dept Artificial Intelligence, Daegu, South Korea | Darlan, Daison/KQA-9542-2024; AJANI, Oladayo/HIR-9607-2022; Mallipeddi, Rammohan/AAL-5306-2020 | 58164208500; 57465126000; 25639919900 | daisondarlan33@gmail.com;oladayosolomon@gmail.com;mallipeddi.ram@gmail.com; | 2023 INTERNATIONAL CONFERENCE ON MACHINE INTELLIGENCE FOR GEOANALYTICS AND REMOTE SENSING, MIGARS | 2.61 | 2025-06-25 | 1 | 3 | Landing Site Selection; Clustering | Clustering; Landing Site Selection | Interplanetary flight; Landing; Learning algorithms; Lunar missions; Machine learning; Planets; Space research; Clusterings; Conventional methods; Engineering constraints; Landing mission; Landing site; Landing site selection; Machine-learning; Manual analysis; Multi-domains; Planetary landing; Site selection | English | 2023 | 2023 | 10.1109/migars57353.2023.10064571 | 바로가기 | 바로가기 | 바로가기 | |||||||||||||
| ○ | Book chapter | Machine learning for renewable energy applications | [No abstract available] | Arumugam, Dhanasekaran; Stephen, Christopher; Parmar, Richa; Jegadeesan, Vishnupriyan; Paul, Ajay John; Kariyama, Ibrahim Denka | Center for Energy Research, Department of Mechanical Engineering, Chennai Institute of Technology, Chennai, Tamil Nadu, India; Department of Mechanical Engineering, Vel Tech Rangarajan Dr. Sagunthala, R & D Institute of Science and Technology, Chennai, Tamil Nadu, India, Teaching Associateship for Research, Excellence (TARE) Fellow under SERB, National Institute of Solar Energy (NISE), Gurugram, Haryana, India; Solar Water Pump Laboratory, National Institute of Solar Energy, Gurugram, Haryana, India; Center for Energy Research, Department of Electrical and Electronics Engineering, Chennai Institute of Technology, Chennai, Tamil Nadu, India; School of Mechanical Engineering, Kyungpook National University, Daegu, Gyeongbuk Province, South Korea; Department of Agricultural Engineering, Dr. Hilla Limann Technical University, Wa, Upper West, Ghana | 57212452016; 58336427100; 57216739022; 56422530100; 58338427000; 48561539000 | Renewable Energy and AI for Sustainable Development | 0 | 2025-06-25 | 0 | English | Final | 2023 | 10.1201/9781003369554-8 | 바로가기 | 바로가기 | |||||||||||||||||||||
| ○ | ○ | Proceedings Paper | Machine Learning-Based Batch Processing for Calibration of Model and Noise Parameters | Non-Gaussian or non-whiteness of noise sources often occurs in many digital avionics systems. Incorrect modeling of the system degrades the performance of parametric model-based estimators and controllers. To calibrate the model and noise parameters, this paper proposes a machine learning-based batch processing approach. We first mathematically formulate a state augmentation system containing three types of noise: color noise, state-dependent noise, and correlation noise. Next, we define accessible process and measurement residuals to create the training data set. Finally, we propose offline batch processing that recursively utilizes a machine learning technique to calibrate the model and noise parameters. Simulation results under various conditions validate the calibration performance of the proposed approach. | Lee, Kyuman | Kyungpook Natl Univ, Dept Robot & Smart Syst Engn, Daegu, South Korea | lee, kyuman/AAM-6979-2020 | 57193932345 | klee400@knu.ac.kr; | 2023 IEEE/AIAA 42ND DIGITAL AVIONICS SYSTEMS CONFERENCE, DASC | 2155-7195 | 0 | 2025-06-25 | 0 | 0 | non-whiteness noise; modeling error; calibration; batch processing; machine learning | batch processing; calibration; machine learning; modeling error; non-whiteness noise | Batch data processing; Digital avionics; Machine learning; Avionic systems; Batch processing; Machine-learning; Model errors; Modeling parameters; Noise parameters; Noise source; Non-Gaussian; Non-whiteness noise; Performance; Calibration | English | 2023 | 2023 | 10.1109/dasc58513.2023.10311101 | 바로가기 | 바로가기 | 바로가기 | ||||||||||||
| ○ | ○ | Proceedings Paper | MAFD: A Federated Distillation Approach with Multi-head Attention for Recommendation Tasks | The key challenges that recommendation systems must overcome are data isolation and privacy protection issues. Federated learning can efficiently train global models using decentralized data while preserving privacy. In real-world applications, however, it is difficult to achieve high prediction accuracy due to the heterogeneity of devices, the lack of data, and the limited generalization capacity of models. In this research, we introduce a personalized federated knowledge distillation model for a recommendation system based on a multi-head attention mechanism for recommendation systems. Specifically, we first employ federated distillation to improve the performance of student models and introduce a multi-head attention mechanism to enhance user encoding information. Next, we incorporate Wasserstein distance into the objective function of combined distillation to reduce the distribution gap between teacher and student networks and also use an adaptive learning rate technique to enhance convergence. We show that the proposed approach achieves better effectiveness and robustness through benchmarks. | Wu, Aming; Kwon, Young-Woo | Kyungpook Natl Univ, Daegu, South Korea | Kwon, Young-Woo/HGE-6607-2022 | 58262125900; 57208480210 | wuaming@knu.ac.kr;ywkwon@knu.ac.kr; | 38TH ANNUAL ACM SYMPOSIUM ON APPLIED COMPUTING, SAC 2023 | 0 | 2025-06-25 | 1 | 0 | Federated learning; Multi-head attention; Wasserstein distance; Recommendation systems | federated learning; multi-head attention; recommendation systems; wasserstein distance | Distillation; Learning systems; Privacy-preserving techniques; Attention mechanisms; Decentralised; Federated learning; Generalization capacity; Global models; Multi-head attention; Prediction accuracy; Privacy protection; Real-world; Wasserstein distance; Recommender systems | English | 2023 | 2023 | 10.1145/3555776.3577849 | 바로가기 | 바로가기 | 바로가기 |
페이지 이동: