http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
위경수(Kyoung-Soo We),박수환(Soo-Hwan Park),이창건(Chang-Gun Lee),하은용(Eun-Yong Ha) 한국정보과학회 2009 한국정보과학회 학술발표논문집 Vol.36 No.1
로봇 산업이 발전하여 보다 정교한 로봇이 등장하면서 로봇 소프트웨어의 복잡성이 증가하고 있다. 그리고 그러한 복잡성을 완화시키기 위해 소프트웨어를 더 작은 모듈로 분리하는 등의 여러 가지 기법들이 제안되고 있다. 하지만 이렇게 로봇 소프트웨어의 복잡성을 낮춤으로써 그에 따른 여러 문제점들도 파생되고 있다. 여기에서는 이러한 문제점들을 로봇 태스크의 특성을 고려하여 미들웨어 레벨에서 처리할 수 있는 전략을 소개한다.
문제 공간 탐색을 통한 머신러닝 기반 오토터레인 로직 개선 개발
위경수(Kyoungsoo We),구정모(JungMo Koo),이혜연(Hye Yeon Lee),이승태(Seung Tae Lee) 한국자동차공학회 2023 한국자동차공학회 부문종합 학술대회 Vol.2023 No.5
Auto-Terrain logic was released in 2021 to enhance controllability of sport utility vehicles on every road surfaces such as paved, snowy, sandy and unpaved ways. For this, this logic uses machine learning techniques to recognize the road surface the car is driving on and automatically selects the appropriate driving mode. To improve the accuracy of the road surface recognition, post-processing has been added to the result of the learning model, and this causes the additional delay in recognizing the road surface. In this paper, to reduce the delay of the post-processing, we try to increase the accuracy of the learning model. For this, we explore the problem space which consists of various input combinations of the model. As a result, we get the improved model which increases the accuracy from 84.4% to 98.5%, and also get the efficient model which reduces the memory usage while minimizing the decrease in accuracy compared to the improved model. Also, this efficient model reduces the post-processing delay by 57% compared to the original model.
대기행렬 모델을 이용한 국내 대규모 공동주택 기초 콘크리트 타설공사 자원 관리 개선에 관한 연구
위경수(Wee, Kyung-Soo),노석호(Noh, Seok-Ho),함남혁(Ham, Nam-Hyuk),김재준(Kim, Jae-Jun) 대한건축학회 2021 대한건축학회논문집 Vol.37 No.9
Resource management is essential in a construction project. In particular, resource leveling, which reduces resource fluctuations among resource management methods, prevents resource stagnation and shortage, thereby maintaining a balance between limited construction cost and construction period. In this study, the effect of resource leveling through the queueing model on the construction cost and construction period was quantitatively analyzed based on the data of the case project. First, in this study, the resource-related problems of the case project are described and the state of the queue system is analyzed using the queueing model. Second, the queue system status is analyzed after resource leveling by objective criteria. Third, quantitatively analyze the effect of resource leveling using the queueing model on construction cost and construction period. Through this, the method proposed in this study is verified. This result helps objective decision making in the process planning stage or resource input decision stage, taking into account waiting costs and service costs.
유동 데드라인 태스크 시스템에서의 효율적 스케줄링 기법
위경수(Kyoung-Soo We),박수환(Soo-Hwan Park),이창건(Chang-Gun Lee) 한국정보과학회 2011 정보과학회 컴퓨팅의 실제 논문지 Vol.17 No.2
실시간 시스템에서는 일반적으로 태스크의 데드라인이 고정되어 있다고 가정하고 최악 응답 시간 분석 기법(Worst-case response time analysis)을 이용하여 스케줄 가능성 테스트(Schedulability check)를 한다. 하지만 고정된 데드라인으로 태스크를 기술하는 방식은 특정 시스템에서는 너무 비관적인 결과를 야기할 수 있다. 이러한 시스템에서는 새로운 방식으로 데드라인을 기술하고 그에 따라 태스크를 스케줄링 하는 방식도 기존의 것과는 달리 하는 것이 더 효율적이다. 본 논문에서는 태스크의 수행 순서에 따라 데드라인이 변하는 상황을 모델링하고 그 상황에서의 효율적인 스케줄링을 위한 기법을 소개한다. Worst-case response time analysis is used for schedulability check in multi-task real-time systems. In general, tasks have fixed deadlines in the system. But this analysis using fixed deadlines may underutilize some specific systems. So new concepts of deadline description, and proper scheduling strategies are needed in such systems. In this paper, for a system where deadlines of tasks are shifting in task execution order, and we propose an effective scheduling strategy for the system.
위경수(Kyoungsoo We),구정모(JungMo Koo),이혜연(Hye Yeon Lee),이승태(Seung Tae Lee) 한국자동차공학회 2023 한국자동차공학회 학술대회 및 전시회 Vol.2023 No.11
In the automotive field, identification of new failures is very important in terms of safety. However, existing techniques have limitations to identify new failures. Therefore, in this paper, we propose an unsupervised learning-based method to detect new failures that cause abnormal driving maneuver. We show that the proposed approach would be implemented with up to 2,000 times less memory while achieving higher accuracy compared to existing approaches.
CPS의 점진적인 개발 과정을 지원하는 실시간 시뮬레이션 프레임워크
한재화,위경수,이창건,Han, Jae-Hwa,We, Kyoung-Soo,Lee, Chang-Gun 대한임베디드공학회 2012 대한임베디드공학회논문지 Vol.7 No.6
When developing a CPS, since it is nature of CPS to interact with a physical system, CPS should be verified during its development process by real-time simulation supporting timely interactions between the simulator and existing implemented hardwares. Furthermore, when a part of a simulated system is implemented to real hardwares, i.e., incremental development, the simulator should aware changes of the simulated system and apply it automatically without manual description of the changes for effective development. For this, we suggest a real-time simulation framework including the concept of 'port' which abstracts communication details between the tasks, and a scheduling algorithm for guaranteeing 'real-time correctness' of the simulator.