• Journal of KIISE:Software and Applications
• /
• v.35 no.3
• /
• pp.179-188
• /
• 2008

An important issue in multiagent reinforcement learning is how an agent should team its optimal policy through trial-and-error interactions in a dynamic environment where there exist other agents able to influence its own performance. Most previous works for multiagent reinforcement teaming tend to apply single-agent reinforcement learning techniques without any extensions or are based upon some unrealistic assumptions even though they build and use explicit models of other agents. In this paper, basic concepts that constitute the common foundation of multiagent reinforcement learning techniques are first formulated, and then, based on these concepts, previous works are compared in terms of characteristics and limitations. After that, a policy model of the opponent agent and a new multiagent reinforcement learning method using this model are introduced. Unlike previous works, the proposed multiagent reinforcement learning method utilize a policy model instead of the Q function model of the opponent agent. Moreover, this learning method can improve learning efficiency by using a simpler one than other richer but time-consuming policy models such as Finite State Machines(FSM) and Markov chains. In this paper. the Cat and Mouse game is introduced as an adversarial multiagent environment. And effectiveness of the proposed multiagent reinforcement learning method is analyzed through experiments using this game as testbed.

• Journal of Satellite, Information and Communications
• /
• v.2 no.2
• /
• pp.16-21
• /
• 2007

Various perturbations by the sun, the moon and the earth itself cause a continuous change in nominal position of a geostationary satellite. In order to maintain the satellite within a required window, north-south station keeping for controlling inclination and right ascension of ascending node, and east-west station keeping for controlling eccentricity and longitude are required. In this paper, station keeping maneuver simulation for Communication, Ocean and Meteorological Satellite (COMS) was performed using COMS Flight Dynamics Software(FDS) and the results were analyzed. COMS performs weekly based east-west/north-south station keeping to maintain satellite within ${\pm}0.05^{\circ}$ at the nominal longitude of $128.2^{\circ}E$. In addition, COMS performs wheel off-loading maneuver twice a day to eliminate attitude error caused by one-solar wing in the south panel of the satellite. In this paper, station keeping maneuver considering wheel off-loading maneuver was performed and the results showed that COMS can be maintained well within ${\pm}0.05^{\circ}$ window using COMS FDS.

• Journal of Korean Academy of Oral and Maxillofacial Radiology
• /
• v.27 no.1
• /
• pp.55-71
• /
• 1997

Digora/sup (R)/ an intraoral digital radiography system utilizing image plate (IP) - has a dynamic range of exposure time which allows it to decrease the patient's exposure time and to increase diagnostic ability through image processing, transmission and storage. The purpose of this study was to evaluate the Digora/sup (R)/ system by assessing the effects of various exposure times on the detectability on the tip of the endodontic file. Examining the root canals of 45 extracted sound premolars, K -files No. 10, 15, and 20 were placed at slightly varying distances from the apex. The teeth were glued onto resin-plaster blocks. Five exposure times varying between 0.01 seconds and 0.25 seconds were used. Four observers were asked to measure the distance between the tip of the file and a reduction of crown portion, and obtained mean errors (subtracting true file length from the measured file length), comparing Digora/sup (R)/ monitors with E-plus films, which were both obtained under the same geometrical positions. The results were as follows : 1. Comparing E-plus film with Digora/sup (R)/ at 0.01 seconds, the mean errors in E-plus film showed -4.453 nun, -4.497 nun, and -3.857 nun, while the mean errors in Digora/sup (R)/ showed 0.065 nun, 0.607 nun, and 0.719 mm according to the file groups. Therefore there was a significant difference between E-plus film and Digora/sup (R)/(p<0.05). 2. By comparison of mean errors according to the various exposure times in the Digora/sup (R)/ system, the mean error at 0.01 seconds was significantly lower than that at 0.12 and 0.25 seconds in the No. 10 file group(p<0.05). And the standard deviation was the highest at 0.01 seconds. 3. Comparing E-plus film at 0.25 seconds with the Digora/sup (R)/ system, the mean errors showed a significant difference between E-plus film at 0.25 seconds and the Digora/sup (R)/ system at 0.25 seconds in No. 10 and 20 file groups(p<0.05). 4. Comparing E -plus film at 0.25 seconds with other exposure times, the mean errors showed a significant difference between E-plus film at 0.25 seconds and E-plus film at 0 .. 01 and 0.03 seconds in 10 file group(p<0.05). In the No. 15 and 20 file groups, there was a significant difference between E-plus film at 0.25 seconds and E-plus film at 0.01 seconds(p<0.05). In conclusion, Digora/sup (R)/ was better than E-plus film in detectability on the tip of the file at the exposure time of 0.01 seconds in all file groups. And we concluded that Digora/sup (R)/ can shorten exposure times up to 4％ of 0.25 seconds (0.01 sec), which is adequate exposure time for premolar in E-plus film using No. 15 and 20 files.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.3
• /
• pp.18-31
• /
• 2002

In this study, we introduce a concept of advanced neurofuzzy polynomial networks(ANFPN), a hybrid modeling architecture combining neurofuzzy networks(NFN) and polynomial neural networks(PNN). These networks are highly nonlinear rule-based models. The development of the ANFPN dwells on the technologies of Computational Intelligence(Cl), namely fuzzy sets, neural networks and genetic algorithms. NFN contributes to the formation of the premise part of the rule-based structure of the ANFPN. The consequence part of the ANFPN is designed using PNN. At the premise part of the ANFPN, NFN uses both the simplified fuzzy inference and error back-propagation learning rule. The parameters of the membership functions, learning rates and momentum coefficients are adjusted with the use of genetic optimization. As the consequence structure of ANFPN, PNN is a flexible network architecture whose structure(topology) is developed through learning. In particular, the number of layers and nodes of the PNN are not fixed in advance but is generated in a dynamic way. In this study, we introduce two kinds of ANFPN architectures, namely the basic and the modified one. Here the basic and the modified architecture depend on the number of input variables and the order of polynomial in each layer of PNN structure. Owing to the specific features of two combined architectures, it is possible to consider the nonlinear characteristics of process system and to obtain the better output performance with superb predictive ability. The availability and feasibility of the ANFPN are discussed and illustrated with the aid of two representative numerical examples. The results show that the proposed ANFPN can produce the model with higher accuracy and predictive ability than any other method presented previously.

• Progress in Medical Physics
• /
• v.20 no.3
• /
• pp.174-179
• /
• 2009

To track moving tumor in real time, CyberKnife system imports a technique of the synchrony respiratory tracking system. The fiducial marker which are detectable in X-ray images were demand in CyberKnife Robotic radiosurgery system. It issued as reference markers to locate and track tumor location during patient alignment and treatment delivery. Fiducial marker implantation is an invasive surgical operation that carries a relatively high risk of pneumothorax. Most recently, it was developed a direct lung tumor registration method that does not require the use of fiducials. The purpose of this study is to measure the accuracy of target applying X-sight lung tracking using the Gafchromic film in dynamic moving thorax phantom. The X-sight Lung Tracking quality assurance motion phantom simulates simple respiratory motion of a lung tumor and provides Gafchromic dosimetry film-based test capability at locations inside the phantom corresponding to a typical lung tumor. The total average error for the X-sight Lung Tracking System with a moving target was $0.85{\pm}0.22$ mm. The results were considered reliable and applicable for lung tumor treatment in CyberKnife radiosurgery system. Clinically, breathing patterns of patients may vary during radiation therapy. Therefore, additional studies with a set real patient data are necessary to evaluate the target accuracy for the X-sight Lung Tracking system.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1993.04a
• /
• pp.40-45
• /
• 1993

일반적으로 건물 구조물에 전달되는 기계진동을 감소시키기 위해서 기계와 기초사이에 유연한 방진소자를 삽입하여 기계가진력(exciting force)의 전달 률을 줄인다. 또한 구조물의 고유진동수와 진동원의 가진주파수가 일치할 경 우, 가진주파수를 변화시키거나, 구조물의 동특성을 변화시키는 방법을 사용 한다. 어떠한 방안을 선택하든 효과적이고 정량적인 방진 시스템을 구성하고 구조물의 정확한 진동상태를 예측하기 위해서는 진동원의 가진특성과 구조 물의 동특성에 대한 정보가 요구된다. 일반적으로 방진설계를 위해 필요한 진동원의 가진특성은 제조회사의 사양이나 측정을 통하여 비교적 쉽게 얻을 수 있다. 복합재료, 다양한 경계조건, 복잡한 대형구조물등은 수치해석을 이 용하여 해석적인 방법으로 동특성을 구할 경우, 신뢰성 있는 정보를 얻기에 는 많은 노력이 요구된다. 더우기 현장에서 발생하는 진동문제는 대부분 복 잡하고 시간적으로 시급히 해결해야 하기 때문에 효율적인 절차를 구성하여 구조물의 동특성을 해석하는 방법을 사용할 필요가 있다. 구조물의 동특성은 실험적인 방법을 통하여 구하고 그 외의 필요한 계산들은 해석을 통하여 얻 는 것이 효율적일 수 있다. 실험적 동특성해석은 입력하중에 대한 응답의 크 기와 위상 비를 주파수별로 나타내는 전달함수를 측정하는 방법으로서 가진 장치 및 여러 측정/분석 장비가 필요하며, 철교, 교량, 건물의 철골 및 콘크 리트 슬라브등 다양한 중대형의 구조물을 Signal/Noise비가 좋도록 가진 시 켜야 할 필요성이 있다. 본 연구에서는 이러한 실험적 방법의 현장 적응성과 신뢰성을 확보하기 위해 대형충격기(large impact hammer, max, peak force 약 10000N, time duration 약 20ms)를 제작하고 실험/분석 시스템 및 구조물 의 진동제어를 위한 절차를 Fig.1과 같이 구성하고 이를 철근콘크리트 건물 에 설치한 기계식 주차설비의 진동제어에 적용하였다.force response simulation)를 수행하여 임의의 좌표 공간에 대한 진동수준을 해석적으로 예측할 뿐만 아니라 구조물의 진동제어 를 위한 동적인자를 변경시킬 수 있는 정보를 제공하며 장비를 방진할 경우 신뢰성 있는 전달률을 결정할 수 있다. 실험적으로 철교, 교량이나 건물의 철골구조 및 2층 바닥 등 대,중형의 복잡한 구조물에 대항 동특성을 나타내 는 모빌리티를 결정할 경우 충격 가진 실험이 사용되는 실험장비 측면에서 나 실험을 수행하는 과정이 대체적으로 간편하다. 그러나 이 경우 대상 구조 물을 충분히 가진시킬수 있는 용량의 대형 충격기(large impact hammer)가 필요하게 된다. 이러한 동적실험은 약 길이 61m, 폭 16m의 4경간 교량에 대 하여 동적실험을 수행하여 가능성을 확인하였다. 여기서는 실험실 수준의 평 판모델을 제작하고 실제 현장에서 이루어질 수 있는 진동제어 구조물에 대 한 동적실험 및 FRS를 수행하는 과정과 동일하게 따름으로써 실제 발생할 수 있는 오차나 error를 실험실내의 차원에서 파악하여 진동원을 있는 구조 물에 대한 진동제어기술을 보유하고자 한다. 이용한 해마의 부피측정은 해마경화증 환자의 진단에 있어 육안적인 MR 진단이 어려운 제한된 경우에만 실제적 도움을 줄 수 있는 보조적인 방법으로 생각된다.ofile whereas relaxivity at high field is not affected by τS. On the other hand, the change in τV does not affect low field profile but strongly in fluences on both inflection fie이 and the maximum relaxivity value. The results shows a fluences on both inflection field and the maximum relaxivity value. The results shows

• Korean Journal of Ecology and Environment
• /
• v.46 no.2
• /
• pp.310-318
• /
• 2013

In territorial ecosystem, soil has stored considerable amount of carbon, and it is vulnerable to weakness release much of the carbon to atmosphere. In this study, we have been effort realization and discussion to the error between inter-instruments and measurement methods, time and special variations, gap filling and separation from each source included in soil respiration, used to collect soil respiration data in various ecosystems in Korea. In conclusion, it have to collect calibration data throughout comparison test between methods and instruments because accumulated data from past and accumulating data in present did not calibrated. In predicting change of soil carbon dynamic using the model method, it needs important data such as longterm and short-term data, artificial handling data of major factor, data from various ecosystem, soil texture, soil depth etc. In company with, we should collect highly qualified data through deep consideration of present problems.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.4
• /
• pp.55-66
• /
• 2020

This study developed four algorithms and their associated indices that can quantify and qualify road hazards along roadways. Initially, relevant raw data can be collected from commercial vehicles by camera and DTG. Well-processed data, such as potholes, road freezing, and fog, can be generated from the Integrated management system. Road hazard algorithms combine these data with road inventory data in the Data Sharing Platform. Depending on well-processed data, four different road hazard algorithms and their associated indices were developed. To test the algorithms, an experimental plan based on passive DTG attached in probe vehicles was performed at two different test locations. Selection of the test routes was based on historical data. Although there were limitations using random data for commercial vehicles, hazardous roadways sections, such as fog, road freezing, and potholes, were generated based on actual historical data. As a result, no algorithm error was found in the entire test. Because this study provides road hazard information according to a section, not a point, it can be practically helpful to road users as well as road agencies.

• Journal of Korea Port Economic Association
• /
• v.28 no.2
• /
• pp.75-93
• /
• 2012

The steel cargoes as the core raw materials for the manufacturing industry have important roles for increasing the handling volume of the port. In particular, steel cargoes are fundamental to vitalize Port of Incheon because they have recognized as the primary key cargo items among the bulk cargoes. In this respect, the IPA(Incheon Port Authority) ambitiously developed the port complex facilities including dedicated terminals and its hinterland in northern part of Incheon. However, these complex area has suffered from low cargo handling records and has faced operational difficulties due to decreased net profits. In general, the import and export steel cargo volumes are sensitively fluctuated followed by internal and external economy index. There is a scant of research for forecasting the steel cargo volume in Incheon port which used in various economy index. To fill the research gap, the aim of this research is to predict the steel cargoes of Port of Incheon using the well established methodology i.e. System Dynamics. As a result, steel cargoes volume dealt with in Incheon port is forecasted from about 8 million tons to about 10 million tons during simulation duration (2011-2020). The Mean Absolute Percentage Error (MAPE) is measured as 0.0013 which verifies the model's accuracy.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1995.10a
• /
• pp.289-295
• /
• 1995

구조물의 동적부하에 대한 동적변형 응답을 정확히 예측하고, Over Design이나 Under Design이 아닌 합리적인 설계방안의 개발은 중요한 과제이다. 동적강도해석이나 소음 승차감과 같은 진동 및 충격에 기인하는 제반 문제를 복잡한 구조물을 대상으로 합리적으로 처리하기 위한 Dynamic Design Analysis는 높은 신뢰성의 추구와 더불어 필요불가결한 기술이 되고 있다. 동적해석 방법으로는 현재 유한요소법이 널리 사용되고 있으며 여러 종류의 범용 프로그램들이 보급되어 있는 실정이다. 그러나 특히 동적문제에 있어서는 형상이나 거동이 복잡한 구조물의 경우, 또는 차량의 차체와 같이 많은 장착물이 부착된 경우에는 유한요소법의 적용이 곤란하여, 지금까지 대처할 수 있는 유용한 방법이 없었다. 따라서 비교적 용이하고 간단하게 적용가능한 진동실험을 기초로 한 구조물의 동적 응답해석 및 설계 방안의 개발이 필요하다. 본 연구에서는 진동시험으로 얻어진 부분구조물의 응답특성과 결합특성으로부터 결합 후의 응답특성을 예측할 수 있는 방법을 전달함수합성이론을 기초로하여 프로그래밍 package화 한다. 그리고 평판구조물에 대하여 진동시험과 컴퓨터 시뮬레이션을 통하여 개발된 방법의 타당성을 검증한다. 또한 실제 차량에서 차체만의 진동시험과 엔진의 자유진동시험에 의한 시험데이터로부터 차체와 엔진이 마운트 결합된 후의 진동특성을 예측한다. 진동시험시에 입력과 출력에 노이즈가 필연적으로 혼입되어 주파수응답함수의 크기(magnitude)와 위상(phase)을 왜곡시킨다. 특히 위상의 왜곡은 복소수연산을 하는 전달함수합성법의 결과에 중요한 영향을 미치게 된다. 본 연구에서는 데이타 획득시 입력과 출력의 시간지연으로 생기는 위상왜곡을 보정하는 방법을 제시하고, 그 개선 정도를 조사한다.는 소견의 확실도로서 가능성을 표현한 것이다. 예를 들면, 진동진폭 스펙트럼상에 2X 성분이 상당히 크게 나타나 정렬불량의 가능성이 0.7 정도라고 판정하는 것 등은 이러한 수치적진리치를 이용하는 방법이다. 그러나 상기의 수치적 표현만으로는 확실도를 한개의 수치로서 대표하게 하는 것은 진단의 정밀도에 문제가 있을 것으로 생각된다. 따라서 언어적진리치가 도입되어 [상당히 확실], [확실], [약간 확실] 등의 언어적인 표현을 이용하여 애매성을 표현하게 되었다. 본 논문에서는 간이진단 결과로부터 추출된 애매한 진단결과중에서 가장 가능성이 높은 이상원인을 복수로 선정하고, 여러 종류의 수치화할 수 없는 언어적(linguistic)인 정보ㄷㄹ을 if-then 형식의 퍼지추론으로 종합하는 회전기계의 이상진단을 위한 정밀진단 알고리즘을 제안하고 그 유용성을 검토한다. 존재하여도 모우드 변수들을 항상 정확하게 구할 수 있으며, 또한 알고리즘의 안정성이 보장된 것이다.. 여기서는 실험실 수준의 평 판모델을 제작하고 실제 현장에서 이루어질 수 있는 진동제어 구조물에 대 한 동적실험 및 FRS를 수행하는 과정과 동일하게 따름으로써 실제 발생할 수 있는 오차나 error를 실험실내의 차원에서 파악하여 진동원을 있는 구조 물에 대한 진동제어기술을 보유하고자 한다. 이용한 해마의 부피측정은 해마경화증 환자의 진단에 있어 육안적인 MR 진단이 어려운 제한된 경우에만 실제적 도움을 줄 수 있는 보조적인 방법으로 생각된다.ofile whereas relaxivity at high field is not affected by τS. On the other hand, the change in τV does not affect low field profile but strongly in fluences on bot