A5038547880- Magnetic confinement fusion research
- Superconducting Materials and Applications
- Particle accelerators and beam dynamics
- Advanced Data Storage Technologies
- Laser-Plasma Interactions and Diagnostics
- Distributed and Parallel Computing Systems
- Robotics and Sensor-Based Localization
- Advanced Image and Video Retrieval Techniques
- Advanced Vision and Imaging
- Fault Detection and Control Systems
- Ionosphere and magnetosphere dynamics
- IPv6, Mobility, Handover, Networks, Security
- Image and Object Detection Techniques
- Energy Load and Power Forecasting
- Medical Imaging Techniques and Applications
- Risk and Safety Analysis
- Meteorological Phenomena and Simulations
- Video Surveillance and Tracking Methods
- EEG and Brain-Computer Interfaces
- Network Traffic and Congestion Control
- Fuzzy Logic and Control Systems
- Context-Aware Activity Recognition Systems
- Anomaly Detection Techniques and Applications
- Educational Robotics and Engineering
- Network Time Synchronization Technologies
Inje University Ilsan Paik Hospital
2025
Korea Institute of Fusion Energy
2016-2024
Korea Advanced Institute of Science and Technology
2010-2013
Heart rate variability (HRV) is a physiological marker of the cardiac autonomic modulation and related emotional regulation. Electroencephalography (EEG) reflective brain cortical activities psychopathology. The HRV EEG have been employed in machine learning- deep learning-based algorithms either alone or with other wearable device-based features to classify patients psychiatric disorder (PT) healthy controls (HC). Little study examined utility markers discern PT various diagnosis versus HC....
To achieve nuclear fusion energy generation, continuous confinement of high-current and long-pulse plasma is necessary. KSTAR has been conducting experiments to this goal. In order maintain a steady state current, the superconducting magnets tokamak should also be operated stably, with heat controlled minimum. However, slight increase in outlet temperature PF (poloidal field) magnet current ripple were observed, particularly when ramp rate 0.5 kA/s or less. range, evaluated AC loss showed...
Abstract Superconducting coils play a critical role in superconducting-based nuclear fusion device. As the temperature of superconducting magnets increases with change current, it is important to predict their prevent excessive rise and operate them efficiently. We present multi-scale recurrent transformer system, deep learning model for forecasting coils. Our system recurrently predicts future data coil using previous obtained from Korea Tokamak Advanced Research poloidal field dataset...
For the systematic standardization of real time control at Korea Superconducting Tokamak Advanced Research (KSTAR) facility, we developed a new functional digital controller based on MTCA.4 Standard. The KSTAR Multifunction Control Unit (KMCU, K-Z35) is realized using Xilinx System-On-Chip (SOC) architecture. KMCU matched with dedicated Rear Transition Module (RTM) sites for two FMC-like analog Data Acquisition (DAQ) modules. first DAQ system to be implemented Motional Stark Effect (MSE)...
Most of the previous visual odometry methods cannot deal with a large independently moving object that takes up over 50% image area. To overcome this problem, spatiotemporal filter is incorporated into RANSAC method to out false match occurrs by object. This uses current and motion vector's length direction. Experimental results demonstrate proposed effectively rejects vectors generated from objects improves accuracy.
This paper describes a visual odometry algorithm that deals with the nearly degenerated situation caused by false motion vector generated independently moving objects, repetitive patterns and wrong depth information often arise in for outdoor service robots. To filter out these vectors, we use temporal spatial filter. The uses previous models to abruptly changed vector's length direction information. of vectors objects are different from camera movement 3D space, triangulation error is...
In this paper, we propose S/W framework for health care robot based on OSGI middleware. Most healthcare developed independently. Therefore most module duplicately developed. However It`s not efficient. paper that can provide common and apply heterogeneous robot. By using framework, enhance reusability of s/w flexible efficient development environment.
This paper proposes a novel method for detection of hand raising poses from images acquired single camera attached to mobile robot that navigates unknown dynamic environments. Due unconstrained illumination, high level variance in human appearances and unpredictable backgrounds, detecting gestures an image is very challenging. The proposed first detects faces determine the region interest (ROI), this ROI, we detect hands by using HOG-based detector. By color distribution face region,...
To standardize and simplify the control system at Korea Superconducting Tokamak Advanced Research (KSTAR), we develop 10G Ethernet based remote DAQ system. By separating host, structure of can be made more flexible. We have developed a module with interface on MicroTCA.4 designed to devices in real time server via 10GE. connect proposed device use network UDP multicast atop 10GbE cut-through packet switching infrastructure. This is implemented using Zynq board, matched RTM board that has...