• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.927-930
• /
• 2005

ACSR is a timed process algebra for the specification and analysis of real-time systems, which supports synchronous timed actions and asynchronous instantaneous events. PACSR is an extended ACSR with the notion of probabilities in selection operation. Using PACSR, this paper represents a system fault occurrence and recovery from the fault in the general resource alteration system. The result shows that system fault occurrence can be analyzed from the fault occurrence probability and the recovery probability.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06b
• /
• pp.106-108
• /
• 2006

강화 학습은 누적 보상 값을 최대화할 수 있는 행동 선택 전략을 학습하는 온라인 학습의 한 형태이다. 효과적인 강화학습을 위해 학습 에이전트가 매 순간 고민해야 하는 문제가 탐험(exploitation)과 탐색(exploration)의 문제이다. 경험과 학습이 충분치 않은 상태의 에이전트는 어느 정도의 보상 값을 보장하는 과거에 경험한 행동을 선택하느냐 아니면 보상 값을 예측할 수 없는 새로운 행동을 시도해봄으로써 학습의 폭을 넓힐 것이냐를 고민하게 된다. 특히 단일 에이전트에 비해 상태공간과 행동공간이 더욱 커지는 다중 에이전트 시스템의 경우, 효과적인 강화학습을 위해서는 상태 공간 축소방법과 더불어 탐색의 기회가 많은 행동 선택 전략이 마련되어야 한다. 본 논문에서는 로봇축구 Keepaway를 위한 효율적인 다중 에이전트 강화학습 방법을 설명한다. 이 방법의 특징은 상태 공간 축소를 위해 함수근사방법의 하나인 타일 코딩을 적용하였고, 다양한 행동 선택을 위해 룰렛 휠 선택 전략을 적용한 것이다. 본 논문에서는 이 방법의 효과를 입증하기 위한 실험결과를 소개한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10b
• /
• pp.137-139
• /
• 2000

이동로봇을 위한 제어기를 개발하려는 폭넓은 연구가 진행되어 왔다. 특히, 몇몇 연구가들은 유전자 알고리즘이나 유전자 프로그래밍과 같은 진화 알고리즘을 사용하여 장애물 피하기, 포식자 피하기, 이동하는 먹이 잡기 등의 기능을 수행하는 이동로봇 제어기를 개발하였다. 이러한 연구 선상에서, 우리는 이동로봇을 제어하기 위해 셀룰라 오토마타 상에서 진화된 CAM-Brain을 적용하는 방법을 보여왔다. 그러나, 이러한 접근방법은 로봇이 복잡한 환경에서 적합한 행동을 수행하도록 만드는데 한계가 있었다. 본 논문에서는, Maes의 행동선택 방법론을 이용하여 간단한 행동을 하도록 진화된 모듈들을 결합함으로써 이러한 문제를 해결하려고 한다. 실험 결과는 이러한 접근방법이 복잡한 환경을 위한 신경망 제어기를 개발하는데 가능성이 있음을 보여주었다.

• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2387-2389
• /
• 2003

Acrobat은 대표석인 비선형, underactuated 시스템이며, acrobot의 제어목적에는 swing-up 제어와 balancing 제어가 있다. 이 두 가지 제어목적을 달성하기 위해 기존에 많은 연구가 진행되었다. 그러나 이 방법들은 두 개의 독립적인 제어기를 acrobot의 상태에 따라 전환하여 사용하는 방법으로서 전환 시점의 선정기준에 대한 어려움과 두 가지 제어목적의 달성을 위한 전체 학습 시간지연의 문제점이 있다. 이를 개선하기 위하여 우리는 acrobot의 두 가지 제어목적을 동시에 해결할 수 있도록 기존에 연구하였던 연속적인 상태공간의 근사화가 가능한 영역기반 Q-학습(Region-based Q-Learning)[11]을 기반으로 한 하나의 제어기로 구현하는 방법을 연구하였다. 제안한 방법을 제작한 acrobot에 적용한 실험을 통하여 그 유용성을 검증하였다.

• Journal of the Society of Disaster Information
• /
• v.6 no.1
• /
• pp.46-59
• /
• 2010

In this study, the safety training of comparative analysis of the realities of Korea's safety training and international experience and practical training for the safety experience of a virtual reality simulator, the development of safe conduct as a controlled motion simulator system, image H / W and the control system works, sound effects H / W and the control system works, 4D special effects (smoke, heat, wind, vibration) and a control system integration, mission control system for the selection and evaluation of the proposal, and safety training on Game S / W of development as we have never experienced an earthquake action plan and evacuate to escape the power of experience and the experience of an earthquake (vibration + video), Also the collapse and a fire escape on the experience of following second disaster, the building collapsed during an escape experience in the field, in case of fire According to the initial fire suppression and fire extinguisher usage experience - experience of smoke and heat to escape in, Moreover, the Daegu subway fire in public places such as subway and evacuated to escape the experience, considering the suggested Simulator.

• Smart Media Journal
• /
• v.13 no.2
• /
• pp.68-84
• /
• 2024

This paper describes the development of a system and algorithms for high-quality welfare services by recognizing behavior development indicators (activity, sociability, danger) in children aged 0 to 2 years old using action recognition technology. Action recognition targeted 11 behaviors from lying down in 0-year-olds to jumping in 2-year-olds, using data directly obtained from actual videos provided for research purposes by three nurseries in the Gwangju and Jeonnam regions. A dataset of 1,867 actions from 425 clip videos was built for these 11 behaviors, achieving an average recognition accuracy of 97.4%. Additionally, for real-world application, the Edge Video Analyzer (EVA), a behavior analysis device, was developed and implemented with a region-specific random frame selection-based PoseC3D algorithm, capable of recognizing actions in real-time for up to 30 people in four-channel videos. The developed system was installed in three nurseries, tested by ten childcare teachers over a month, and evaluated through surveys, resulting in a perceived accuracy of 91 points and a service satisfaction score of 94 points.

• Clinical and Experimental Reproductive Medicine
• /
• v.25 no.1
• /
• pp.93-102
• /
• 1998

Normal and abnormal follicular growth and steroidogenesis depend on gonadotropins as well as intraovarian peptides, which may mediate or potentiate gonadotropin action. Inhibin also affect follicular development and steroidogenesis and may play a role in dominant follicle selection and follicular atresia. Therefore, we studied the differences of serum inhibin, gonadotropin and androgen levels in the women with only the ultrasound findings and no disorder, and polycystic ovary (PCO) with ovulatory disturbance. We prospectively analysed forty-three women with PCO. The diagnosis of PCO was based on typical appearance of the ovaries on TVS. Twelve women with regular menstrual cycle and normal ovarian morphology were selected as control. Basal levels of inhibin, luteinizing hormone (LH), follicle stimulating hormone (FSH), estradiol $(E_2)$, testosterone (T), androstenedione (ADD), dehydroepiandrosterone-sulfate (DS), prolactin and TSH in serum were determined. There were significant differences in basal LH levels and LH/FSH ratio between the control and the women with PCO. The basal levels of inhibin and $E_2$ in the oligo-amenorrheic PCO (N=34) were significantly higher than those in the control. There was higher negative correlation between the inhibin and T levels in the oligo-amenorrheic PCO, but, not in the regular cycling PCO. Also, there was higher positive correlation between the LH and T levels in the oligo-amenorrheic PCO, but not in the regular cycling PCO. These data presume that the initial event of PCO is elevated pituitary LH secretion. Elevated levels of LH may down-regulate LH receptors on granulosa cells and also cause hypertrophy of the thecal layer. High level of androgen secreted by the hypertrophied thecal layer may stimulate inhibin secretion from granulosa cells and can be converted to estrogen by extraovarian tissues and could serve to augment pituitary sensitivity to GnRH with a resultant secretion of more LH than FSH. Inhibin may inhibit FSH action on granulosa cell in the PCO follicle, impairing follicular development and dominant follicle selection resulted in ovulatory disturbance.

• Journal of Korea Society of Industrial Information Systems
• /
• v.10 no.2
• /
• pp.1-12
• /
• 2005

The objective of this study is to develop an IS curriculum and its action plan for footwear industry based on the Instructional System Design(ISD) that is widely used in curriculum developments. To this end, six steps, including a goal setting, analysis, design, development, implementation, and evaluation are employed. Firstly, the explicit goal of IS curriculum development is defined and demands of IS education from the footwear companies are identified with questionnaires and interviews. In the design stage, the IS curriculum map is presented. With this map, the specific IT courses are established, classified by a type of industry, a function, and a level of skills. For the implementations, such particular actions as a selection of textbooks, places, managing trainees, and a training support plan are explained. Finally, the various evaluation methods are presented. The suggested IS curriculum can be applied to IS educations for the footwear companies and other related industries.

• International Journal of Industrial Entomology and Biomaterials
• /
• v.9 no.2
• /
• pp.279-281
• /
• 2004

Components of genetic variation were estimated for five metric traits using 24 biparental progenies (N. C. Design III) generated from F$_2$ generation of a commercial bivoltine silkworm hybrid, SH$_{6}$${\times}$NB$_4$D$_2$. Variance due to additive ($\sigma$$^2$A) and dominance ($\sigma$$^2$D) gene effects was significant for single cocoon weight and shell weight. However, magnitude of former was greater than latter indicating preponderance of additive gene action in the inheritance of these two traits. Average degree of dominance was in the range of partial dominance for all the traits. High estimates of heritability (ns) indicated operation of genes with large additive effects, hence, scope exists for improvement of present populations through a few cycles of selection.n.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.3
• /
• pp.9-17
• /
• 2021

In this paper, an autonomous multi-agent path planning using reinforcement learning for monitoring of infrastructures and resources in a computationally distributed system was proposed. Reinforcement-learning-based multi-agent exploratory system in a distributed node enable to evaluate a cumulative reward every action and to provide the optimized knowledge for next available action repeatedly by learning process according to a learning policy. Here, the proposed methods were presented by (a) approach of dynamics-based motion constraints multi-agent path-planning to reduce smaller agent steps toward the given destination(goal), where these agents are able to geographically explore on the environment with initial random-trials versus optimal-trials, (b) approach using agent sub-goal selection to provide more efficient agent exploration(path-planning) to reach the final destination(goal), and (c) approach of reinforcement learning schemes by using the proposed autonomous and asynchronous triggering of agent exploratory phases.