Search

Your search keyword '"Kim, Min Su"' showing total 2,593 results
Start Over Author "Kim, Min Su"
2,593 results on '"Kim, Min Su"'

2. Generative Pre-Training of Time-Series Data for Unsupervised Fault Detection in Semiconductor Manufacturing

Author

Lee, Sewoong, Choi, JinKyou, and Kim, Min Su

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence
Abstract

This paper introduces TRACE-GPT, which stands for Time-seRies Anomaly-detection with Convolutional Embedding and Generative Pre-trained Transformers. TRACE-GPT is designed to pre-train univariate time-series sensor data and detect faults on unlabeled datasets in semiconductor manufacturing. In semiconductor industry, classifying abnormal time-series sensor data from normal data is important because it is directly related to wafer defect. However, small, unlabeled, and even mixed training data without enough anomalies make classification tasks difficult. In this research, we capture features of time-series data with temporal convolutional embedding and Generative Pre-trained Transformer (GPT) to classify abnormal sequences from normal sequences using cross entropy loss. We prove that our model shows better performance than previous unsupervised models with both an open dataset, the University of California Riverside (UCR) time-series classification archive, and the process log of our Chemical Vapor Deposition (CVD) equipment. Our model has the highest F1 score at Equal Error Rate (EER) across all datasets and is only 0.026 below the supervised state-of-the-art baseline on the open dataset.

Published
2023

14. Sensor acquired reachable workspace in the elderly population: A cross-sectional observational study

Author

Chan, Vicky, Thai, Richard, Vartanian, Revik, Kim, Min Su, Hatch, Maya N, Koh, Jason, and Han, Jay J

Subjects
Biomedical and Clinical Sciences, Clinical Sciences, Clinical Trials and Supportive Activities, Clinical Research, Activities of Daily Living, Aged, Cross-Sectional Studies, Humans, Movement, Range of Motion, Articular, Reproducibility of Results, Upper Extremity, elderly, kinect, reachable workspace, upper extremity
Abstract

The elderly population experiences a decline in upper extremity range of motion (ROM), impairing activities of daily living. The primary mode of quantification is by goniometer measurement. In this cross-sectional observation study, we investigate a sensor-acquired reachable workspace for assessing shoulder ROM decline in an elderly population in comparison to traditional measurements. Sixty-one healthy subjects aged ≥ 65 years were included and compared to a cohort of 39 younger subjects, aged 20 to 64. A sensor acquired reachable workspace using a Kinect motion capture camera measured the maximum reaching ability of both arms while in a seated position, measured in m2 and normalized to arm length to calculate a novel score defined as a relative surface area. This score approximates range of motion in the upper extremity. This measurement was compared to goniometer measurements, including active ROM in shoulder flexion and abduction. Total RSA shows moderate to strong correlation between goniometer in flexion and abduction in the dominant arm (R = 0.790 and R = 0.650, P < .001, respectively) and moderate correlations for the nondominant arm (R = 0.622 and R = 0.615, P < .001). Compared to the younger cohort, the elderly population demonstrated significantly reduced total RSA in the dominant arm (meanelderly = 0.774, SD = 0.09; meanyounger = 0.830, SD = 0.07, P < .001), with significant reductions in the upper lateral quadrant in both arms (dominant: meanelderly = 0.225, SD = 0.04; meanyounger = 0.241, SD = 0.01; P < .001; nondominant: meanelderly = 0.213, SD = 0.03; meanyounger = 0.228, SD = 0.01; P = .004). The test-retest reliability was strong for both dominant and nondominant total RSA (ICC > 0.762). The reachable workspace demonstrates promise as a simple and quick tool for clinicians to assess detailed and quantitative active shoulder ROM decline in the elderly population.

Published
2022

15. Dynamic Humanoid Locomotion over Uneven Terrain With Streamlined Perception-Control Pipeline

Author

Lee, Moonyoung, Kwon, Youngsun, Lee, Sebin, Choe, JongHun, Park, Junyong, Jeong, Hyobin, Heo, Yujin, Kim, Min-su, Sungho, Jo, Yoon, Sung-Eui, and Oh, Jun-Ho

Subjects
Computer Science - Robotics
Abstract

Although bipedal locomotion provides the ability to traverse unstructured environments, it requires careful planning and control to safely walk across without falling. This poses an integrated challenge for the robot to perceive, plan, and control its movements, especially with dynamic motions where the robot may have to adapt its swing-leg trajectory onthe-fly in order to safely place its foot on the uneven terrain. In this paper we present an efficient geometric footstep planner and the corresponding walking controller that enables a humanoid robot to dynamically walk across uneven terrain at speeds up to 0.3 m/s. As dynamic locomotion, we refer first to the continuous walking motion without stopping, and second to the on-the-fly replanning of the landing footstep position in middle of the swing phase during the robot gait cycle. This is mainly achieved through the streamlined integration between an efficient sampling-based planner and robust walking controller. The footstep planner is able to generate feasible footsteps within 5 milliseconds, and the controller is able to generate a new corresponding swing leg trajectory as well as the wholebody motion to dynamically balance the robot to the newly updated footsteps. The proposed perception-control pipeline is evaluated and demonstrated with real experiments using a fullscale humanoid to traverse across uneven terrains featured by static stepping stones, dynamically movable stepping stone, or narrow path., Comment: Submitted to Int. Conf. Robotics and Automation (ICRA) 2021

Published
2020

Catalog

Books, media, physical & digital resources
See catalog results