리던던시 환경에서 전기신호식 지능형 조향 시스템의 중재 제어 방법에 관한 연구

김대성(Daesung Kim),김대성(Daesung Kim),이진환(Jinhwan Lee),남궁주(Joo Namgung) 한국자동차공학회 2021 한국자동차공학회 부문종합 학술대회 Vol.2021 No.6

The reliability is described as a probability that a system will operate properly for a defined period of time without any failures, and it is considered as one of the important design attributes. In many safety critical devices, such as the steer-bywire (SbW) in automotive industry, some parts of the control system are considering redundant control system architectures, which are configured two identical controllers in a system, to ensure the higher functionality, reliability and intelligence which can recover from failures. Basically, redundancy is the duplication of critical components or functions of a system with the intention of increasing reliability of the system. Two identically configured controllers (one active, one backup or failsafe) should swap a role in the redundancy scheme within the deadline when a system failure is detected in active controller. A challenge of dynamic redundant systems is to determine a precise role as the active controller to operate a system, error detection of the active controller and how to take over to a backup controller in the defined time. Moreover, a role of each controller has to be independently protected to ensure mutual exclusion under functional safety requirements. This paper studies dynamic redundant architectures and arbitration control methods that can provide full fault-tolerance without any deviation of functionality even in the presence of faults. The evaluation of the proposed redundancy system is performed experimentally by actual test scenarios to show the practical implication of the system architecture.

머신 러닝 기반 엔지니어링 메트릭을 고려한 차량용 임베디드 소프트웨어 결함 예측에 관한 실험적 연구

김대성(Daesung Kim),이성훈(Sunghoon Lee),성기태(Gitae Seong),김대성(Daesung Kim),이진환(Jinhwan Lee),배홍용(Hongyong Bhae) 한국자동차공학회 2020 한국자동차공학회 학술대회 및 전시회 Vol.2020 No.11

The complexity of automotive software has rapidly been increased with vehicular intelligence. Software fault prediction (SFP) can be helpful for improving the software quality and efficiently managing limited testing resources with early identification of faulty module. Many papers have been proposed to predict the potential software fault. Most of them are only considering source code attributes, as a result, it has been remained a challenge for industrial area to adopt which method is fit, robust, and provide most accurate model. In this paper, the framework is presented for SFP model design of the automotive steering system software obtained by mining historical repositories. For best performance model, we have additionally chosen engineering features like difficult level of requirements, competence of engineer, complexity of ports, test coverage with code metrics. Our model is evaluated with AUROC which is a performance measurement for classification problem. The Random Forest Classifier shows the best performance with 92.31%.

압축 감지 기술과 무선통신 응용

황대성(Daesung Hwang),김대성(Daesung Kim),최진호(Jinho Choi),하정석(Jeongseok Ha) 大韓電子工學會 2009 電子工學會論文誌-SP (Signal processing) Vol.46 No.5

Compressed Sensing (이하 압축 감지 기술)은 Nyquist 률 이하로 아날로그 신호를 샘플 할 수 있는 기법이다. 이 기법으로 신호는 기존의 신호 샘플 방법보다 적은 수의 측정값으로 표현이 가능하며 또한 선형 프로그래밍을 이용하여 측정값으로부터 본래 신호를 높은 확률로 복원할 수 있다. 이를 통해 압축 감지 기술은 같은 신호를 획득하는데 소모되는 측정 시간 및 ADC (analog-to-digital converter) 자원의 양을 크게 감소시키는 장점을 갖는다. 본 논문에서는 압축 감지 기술에 대한 기본적인 개념과 임의 기저를 이용하여 아날로그 신호로부터 측정값을 획득하는 방법과 본래 신호를 복원하는 방법에 대해 설명하고, 무선통신 분야에서의 압축 감지 기술 응용 예시를 소개한다. Compressed Sensing is a method to sample analog signals at a rate under the Nyquist rate. With this scheme, it is possible to represent signals with a relatively smaller number of measurements than that of the conventional sampling method, and the original signals are reconstructed with high probability from the acquired measurements using the linear programming. Compressed sensing allows measurement time and/or the amount of ADC (analog-to-digital converter) resources for the signal acquisitions to be reduced. In this paper, we presents the backgrounds of the compressed sensing, a way to acquire measurements from an analog signal with a random basis, and the signal recovery method. Also, we introduce applications of compressed sensing in wireless communications.

ANALYSIS OF THE FIXED BED REACTOR FOR DME SYNTHESIS

Daesung Song,Sung Joon Ahn,Wonjun Cho,Dal Keun Park,En Sup Yoon 제어로봇시스템학회 2006 제어로봇시스템학회 국제학술대회 논문집 Vol.2006 No.10

Net Spillover Effects of Industry Indices’ Return and Volatility

Daesung Jung(정대성),Taehyuk Kim(김태혁),JongHae Park(박종해) 한국산업경제학회 2019 산업경제연구 Vol.32 No.6

본 연구는 Diebold and Yilmax(2012)에 의해 제안된 전이지수를 이용하여 업종간 전이효과와 그 특성을 분석하였다. 주요 실증결과는 다음과 같다. 첫째, 전이지수를 통해서 산업간 수익률전이현상과 변동성전이현상이 관찰되었다. 둘째, 다른 산업에 영향을 주는 크기는 서비스업, 운수창고, 건설업, 제조업, 유통업, 통신업, 전기가스, 금융업 순으로 나타났다. 셋째, 다른 산업으로부터 영향을 받는 크기는 서비스업, 유통업, 제조업, 금융업, 건설업, 운수창고, 전기가스, 통신업 순으로 나타났다. 넷째, 전이지수는 2008년 글로벌 금융위기 때 가장 높게 나타났다. 이상의 결과는 새롭게 제시된 전이지수를 통해서 산업간 전이효과를 살펴보았으며, 이러한 정보는 전이효과를 고려한 새로운 투자전략과 위험관리 측면에서 이용이 가능할 것으로 기대된다. This paper tries to examine spillover effects and traits across sectors of the Korea stock market using the spillover index proposed by Diebold and Yilmaz (2012). Main empirical results found are as follows: 1) By estimating the net spillover index, across sectors, both the spillover of the industry stock returns and the spillover of volatilities are observed. 2) The order of the magnitude of giving spillover effects are as follows; services, transportation and warehouse, construction, manufacturing, distribution, communication, electricity and gas, and finance. 3) The order of the magnitude of giving to others effects of the return is found as follows; KOSPI, manufacturing, distribution, finance, services, transportation and warehouse, electricity and gas, construction, and communication. 4) Based on the estimation of time varying net spillover of the volatility. it is confirmed that KOSPI and manufacturing industries’ volatilities consistently lead other industries. 5) The magnitude of the spillover index shows highest during 2008 global finance crisis. Informations reported here must be useful for new portfolio investment strategy and risk management taking care of spillover effects across sectors.

Estimating Wood Weight Change on Air Drying Times for Three Coniferous Species of South Korea

Daesung Lee,Jungkee Choi 강원대학교 산림과학연구소 2016 Journal of Forest Science Vol.32 No.3

The purposes of this study are to calculate the green and dried weight using wood discs, to figure out weight change on air drying times, and to develop the model of wood disc weight change for Larix kaempferi, Pinus koraiensis, and Pinus densiflora. The variables affecting the weight change were investigated, and the pattern of weight change over time was figured out through linear models. When comparing the stem green weight calculated using wood discs in this study with the weight table of Korea Forest Service, the weight was not significantly different for L. kaempferi and P. koraiensis. On the other hand, in comparison of stem dried weight, the weight was significantly different in all of three species. In addition, various measurement factors were examined to figure out the relationship with weight change, and air drying times and disc diameter were found as significant independent variables. Finally, two linear models were developed to estimate air drying times of three species, fit statistics were significant for practical use.

Estimation of Site Index for Larix kaempferi and Pinus koraiensis in Gangwon and North Gyeongsang Provinces

Daesung Lee,Yeongwan Seo,Gildong Park,Jungkee Choi 강원대학교 산림과학연구소 2015 Journal of Forest Science Vol.31 No.3

Site index curves were developed for Larix kaempferi and Pinus koraiensis in Gangwon and North Gyeongsang provinces in Korea. For the development of site index, Schumacher and Chapman-Richards model were applied using the data collected from 2012 to 2014. Base age was set to 40 years for Larix kaempferi and Pinus koraiensis in site index of this study. Coefficient of determination and root mean square error of site index models were provided by species, and the models were compared with the previous studies to check the suitability. Overall, site index models developed in this study fitted in the current data well. Thus, the site indexes are considered to be properly used in Gangwon and North Gyeongsang provinces.

Synthesis and Characterization of Layer-Patterned Graphene on Ni/Cu Substrate

Daesung Jung,Wooseok Song,Seung Youb Lee,Yooseok Kim,Myoung-Jun Cha,Jumi Cho 한국진공학회 2013 한국진공학회 학술발표회초록집 Vol.2013 No.2

Leaving the Parental Home among Young Adults in Korea

Daesung Choi 한국사회학회 2018 한국사회학회 심포지움 논문집 Vol.2018 No.5

Multi-sensor Fusion for Vehicle Localization in Real Environment

Daesung Jeon,Hoimyung Choi 제어로봇시스템학회 2015 제어로봇시스템학회 국제학술대회 논문집 Vol.2015 No.10

It is essential for vehicles acquiring one’s own current position to navigate autonomously. Especially, vehicles usually do not clearly obtain GPS information at indoor environments. In order to localize itself, vehicles have to use absolute and relative measurement systems from surrounding environmental and its driving information. Absolute position information such as environmental structures and magnets located on the ground can be used for offsetting the error of vehicle’s relative position estimated from encoder and inertial sensors. In this paper, we describe how a variety of multi-sensors are integrated to measure the precise position of a vehicle. This means that as a way to reduce the relative position error, sensor fusion is able to produce accurate navigation information. In order to sense indoor environmental structures such as walls, artificial cones and magnets, we have employed LiDAR sensors and a magnet ruler, a device can detect magnetic field. Thus, we apply these information to Extended Kalman Filter for estimating vehicle’s accurate position, and conduct two experiments to prove the effectiveness of the localization.