• 디자인학연구
• /
• 제20권3호
• /
• pp.97-106
• /
• 2007

본 연구는 교사의 교수 방식에서 효과적으로 사용되는 행동주의 학습이론 중 강화 이론이 교사 보조 로봇에도 효과적으로 적용되는지를 검토하는 데 그 목적이 있다. 피험자내 설계 실험(n=36)으로 성적유형에 따른 우등, 열등 피험자가 강화 유형에 따른 세 가지 로봇의 인터랙션(2*3)을 경험케 했다. 즉, 강화이론과 토큰강화 방식에 기초해 '정적 강화'를 제공하는 로봇('칭찬이'), '부적 강화'를 제공하는 로봇('엄벌이'), '정적 강화'와 '부적 강화'를 모두 제공하는 로봇('상벌이')의 인터랙션을 디자인하고 로봇유형과 피험자의 성적유형에 따른 학생들의 로봇에 대한 인상 및 선호도를 알아보았다. 결과적으로 학생들은 정적 강화를 제공하는 로봇을 가장 선호하고, 부적 강화를 제공하는 로봇을 가장 덜 선호함이 검증되었다. 또한, 강화의 자극을 디자인함에 있어서는 우등 학생에게 부적 강화를 제공하는 로봇에서 로봇이 주는 자극수가 낮을수록 로봇에 대한 긍정적 인상이 증가함을 알 수 있었다. 본 연구 결과는 강화 유형에 따른 학생들의 교사 보조 로봇에 대한 인상 및 선호도를 검증하며, 교사 보조 로봇의 인터랙션 디자인에 효과적인 가이드라인으로 적용될 수 있을 것으로 기대된다.

• 한국구조물진단유지관리공학회 논문집
• /
• 제21권4호
• /
• pp.13-20
• /
• 2017

비보강 조적조 건축물은 재료의 특성상 지진과 같은 횡력에 취약하지만, 국내에는 여전히 많은 조적조 건물이 존재한다. 특히 현재 남아있는 조적조 건축물의 대부분이 20년 이상 노후화됨에 따라 재해감소를 위한 경제성 있는 보강법의 개발이 요구된다. 본 논문에서는 이러한 노후 된 조적조 건축물의 보강법의 하나로 접착형 보강재를 활용한 조적벽체의 외부 보강법을 제시하였으며, 보강효과 검증을 위해 총 6개의 실험체를 형상비(L/H=1.0, 1.3, 2.0)를 변수로 제작하여 정적가력실험을 실시하였다. 실험결과, 보강전 후 조적벽체는 강체회전 및 미끄러짐에 의해 파괴가 발생하였고, 접착형 보강재 부착후 벽체의 최대내력, 최대변위, 소산에너지량은 증가하여 우수한 보강효과를 확인하였다. 또한 기존 유리섬유를 활용한 증가된 전단강도식에 착안하여 비보강 조적벽체에 대한 접착형 보강재의 설계안을 도출하여 적용을 위한 기초자료를 제공하였다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 ISIS 2003
• /
• pp.66-69
• /
• 2003

This paper considers a reinforcement learning(RL) which deals with real environments. Most reinforcement learning studies have been made by simulations because real-environment learning requires large computational cost and much time. Furthermore, it is more difficult to acquire many rewards efficiently in real environments than in virtual ones. The most important requirement to make real-environment learning successful is the appropriate construction of the state space. In this paper, to begin with, I show the basic overview of the reinforcement learning under real environments. Next, 1 introduce a state-space construction method under real environmental which is State Partition Method. Finally I apply this method to a robot navigation problem and compare it with conventional methods.

• 한국해양공학회지
• /
• 제8권1호
• /
• pp.96-104
• /
• 1994

The use of advanced composite materials has grown in recent years in aerospace and other structures. Out of various kinds of repairing methods the one selecteh for this study is an idealized case which simulates a situation where a damaged laminate has been repaired by drilling a hole and therefter plugging the hole with reinforcement. Two typesof reinforcement are investigated ;adhesively bonged plug reinforcement or snug-fit unbonded plug in the hole. For each case of reinforcement, four different sizes of hole diameter and three types of reinforcing material(steel, aluminum, plexiglass) are employed for investigation. The experiment are mainloy forced on the evaluation of ultimate strength of laminate with reinforced hole in comparison to its counterpart with the open hole.

• 한국인터넷방송통신학회논문지
• /
• 제10권4호
• /
• pp.157-162
• /
• 2010

Cart-pole 균형 문제는 유전자 알고리즘, 인공신경망, 강화학습 등을 이용한 제어 전략 분야의 표준 문제이다. 본 논문에서는 cart-pole 균형문제를 해결하기 위해 실시간 강화 학습을 이용한 접근 방법을 제안하였다. 본 논문의 목적은 cart-pole 균형 문제에서 OREL 학습 시스템의 학습 방법을 분석하는데 있다. 실험을 통해, 본 논문에서 제안한 OREL 학습 방법은 Q-학습보다 최적 값 함수에 더 빠르게 접근함을 알 수 있었다.

• International Journal of Precision Engineering and Manufacturing
• /
• 제3권1호
• /
• pp.33-42
• /
• 2002

Recently, various experiments to apply reinforcement learning method to the self-learning intelligent control of continuous dynamic system have been reported in the machine learning related research community. The reports have produced mixed results of some successes and some failures, and show that the success of reinforcement learning method in application to the intelligent control of continuous control systems depends on the ability to combine proper function approximation method with temporal difference methods such as Q-learning and value iteration. One of the difficulties in using function approximation method in connection with temporal difference method is the absence of guarantee for the convergence of the algorithm. This paper provides a proof of convergence of a particular function approximation method based on \"barycentric interpolator\" which is known to be computationally more efficient than multilinear interpolation .

• Carbon letters
• /
• 제7권1호
• /
• pp.27-33
• /
• 2006

Carbon-ceramic composites were fabricated by using fly ash and PANOX fibers as reinforcement. Fly ash, because of its small size particles e.g. submicron to micron level can be effectively dispersed along with fibrous reinforcements. Phenolic resin was used as carbon precursor. Both dry as well as wet methods were used for forming composites. The resulting composites were characterized for their microstructure, thermal and mechanical properties. The microstructure and mechanical properties of composites are found to be dependent on type of the fly ash, fibrous reinforcements as well as processing parameters. The addition of fly ash improves hardness and the fibers, which get co-carbonized on heat treatment, increase the flexural strength of the carbon-ceramic composites. Composites with dual reinforcement exhibit about 30-40% higher strength as compared to the composites made with single reinforcement, either with fly ash as filler or with chopped fibers.

• 자연, 터널 그리고 지하공간
• /
• 제1권1호
• /
• pp.67-80
• /
• 1999

The Grout-reinforcement technique which is widely used during the excavation of a shallow or an endangered tunnel can be classified into a couple of groups according to the properties and injection methods of the grout. The reinforcement design will, therefore, take a different approach based on the grouting method under consideration. However, the injection procedure is mainly performed by the experience of the foreman rather than engineering judgement , specifically the permeation grouting through the rock joints and its reinforcement effect Is not fully under-stood during the design stage, In this study, the anisotropic material properties of the grout-reinforced rock masses are derived from the concept of composite materials and the effect of intact rock, vertical grouting and permeation grouting is, therefore, fully accounted for. Through the parametric studies on the characteristics of rock joints, intact rock and grouting materials, various case studies have been considered. The results, illustrated via the design charts, can be directly used during the reinforcement design.

• Journal of Semiconductor Engineering
• /
• 제2권1호
• /
• pp.125-129
• /
• 2021

As the interest of deep learning, techniques to control the color of images in image processing field are evolving together. However, there is no clear standard for color, and it is not easy to find a way to represent only the color itself like the color-palette. In this paper, we propose a novel color palette extraction system by chroma fine-tuning with reinforcement learning. It helps to recognize the color combination to represent an input image. First, we use RGBY images to create feature maps by transferring the backbone network with well-trained model-weight which is verified at super resolution convolutional neural networks. Second, feature maps are trained to 3 fully connected layers for the color-palette generation with a generative adversarial network (GAN). Third, we use the reinforcement learning method which only changes chroma information of the GAN-output by slightly moving each Y component of YCbCr color gamut of pixel values up and down. The proposed method outperforms existing color palette extraction methods as given the accuracy of 0.9140.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 춘계학술대회
• /
• pp.47-49
• /
• 2022

Robot navigation has seen a major improvement since the the rediscovery of the potential of Artificial Intelligence (AI) and the attention it has garnered in research circles. A notable achievement in the area was Deep Learning (DL) application in computer vision with outstanding daily life applications such as face-recognition, object detection, and more. However, robotics in general still depend on human inputs in certain areas such as localization, navigation, etc. In this paper, we propose a study case of robot navigation based on deep reinforcement technology. We look into the benefits of switching from traditional ROS-based navigation algorithms towards machine learning approaches and methods. We describe the state-of-the-art technology by introducing the concepts of Reinforcement Learning (RL), Deep Learning (DL) and DRL before before focusing on visual navigation based on DRL. The case study preludes further real life deployment in which mobile navigational agent learns to navigate unbeknownst areas.