전자 유압 엑추에이터를 포함하는 비선형 능동 서스펜션의 상태추정을 위한 가우시안 분포 리샘플링 기반 입자 필터 설계

최현덕(Hyun-Duck Choi),유성현(Sung Hyun You) 제어로봇시스템학회 2021 제어·로봇·시스템학회 논문지 Vol.27 No.2

This paper proposes a GDRPF (Gaussian Distribution Resampling Particle Filter) for electro-hydraulic active suspension systems. In order to estimate the state of the nonlinear/non-Gaussian suspension system effectively, a particle filter-based state estimator was developed, and the computational burden and sample impoverishment problems were overcome through the GDR algorithm. The proposed GDRPF-based state estimator not only reduces the computational burden of the particle filters efficiently, but also has better estimate performance than other linearization-based state estimators. Simulation results for nonlinear active suspension system demonstrate the effectiveness of the proposed algorithm.

최적의 측정값 구간의 길이를 갖는 최소 공분산 유한 임펄스 응답 필터 기반 디지털 위상 고정 루프 설계

유성현(Sung-Hyun You),배동성(Dong-Sung Pae),최현덕(Hyun-Duck Choi) 한국전자통신학회 2021 한국전자통신학회 논문지 Vol.16 No.4

디지털 위상 고정 루프는 위상 동기화를 위해 사용되는 회로로 일반적으로 통신, 회로분야 등 다양한 분야에서 사용된다. 디지털 위상 고정 루프를 설계 시 상태추정기를 사용하는 경우 보통 칼만 필터와 같은 무한 임펄스 응답 상태추정기를 활용해왔다. 일반적으로 무한 임펄스 응답 상태추정기 기반 디지털 위상 고정 루프의 성능은 우수하지만, 초기값의 부정확, 모델 오차, 외란 등의 예상하지 못하는 상황에서 급격한 성능저하가 발생할 수 있다. 본 논문에서는 새로운 디지털 위상 고정 루프를 설계 하기 위해 최적의 측정값 구간 길이를 갖는 최소 공분산 유한 임펄스 응답 필터를 제안한다. 제안된 유한 임펄스 응답 필터의 중요 파라미터인 측정값 구간 길이를 구하기 위해 수치적 방법을 소개하며, 필터의 이득을 얻기 위해 비용함수로 오차의 공분산 행렬을 설정하고, 이를 최소화 하기 위하여 선형 행렬 부등식을 사용하였다. 제안된 디지털 위상 동기 루프의 우수성과 강인성을 검증하기 위해 노이즈 정보가 부정확한 상황에서 기존 방법과의 비교 및 분석을 위한 시뮬레이션을 수행하였다. The digital phase-locked loops(DPLL) is a circuit used for phase synchronization and has been generally used in various fields such as communication and circuit fields. State estimators are used to design digital phase-locked loops, and infinite impulse response state estimators such as the well-known Kalman filter have been used. In general, the performance of the infinite impulse response state estimator-based digital phase-locked loop is excellent, but a sudden performance degradation may occur in unexpected situations such as inaccuracy of initial value, model error, and disturbance. In this paper, we propose a minimum variance finite impulse response filter with optimal horizon for designing a new digital phase-locked loop. A numerical method is introduced to obtain the measured value interval length, which is an important parameter of the proposed finite impulse response filter, and to obtain a gain, the covariance matrix of the error is set as a cost function, and a linear matrix inequality is used to minimize it. In order to verify the superiority and robustness of the proposed digital phase-locked loop, a simulation was performed for comparison and analysis with the existing method in a situation where noise information was inaccurate.

유한기억 필터 기반 DPLL 최적설계

유성현(Sung Hyun You),배동성(Dong Sung Pae),최현덕(Hyun Duck Choi) 한국정보기술학회 2023 한국정보기술학회논문지 Vol.21 No.3

The digital phase-locked loop(DPLL) is one of the circuits composed of a digital detector, digital loop filter, voltage controlled oscillator, divider, and etc. as a fundamental circuit, widely used in many fields such as electrical and circuit fields. In order to improve the performance of the DPLL, research has been conducted to improve the performance of the digital loop of the DPLL. An infinite impulse response(IIR) state estimator, which is a mathematical algorithm, is used as one of various methods for improving performance. In this paper, we propose a DPLL based on FIR(Finite Impulse Pulse Response) state estimator. A DPLL using the Frobenius norm-based state estimator, which has more robust performance in inaccurate situations than the minimum variance FIR filter. The Frobenius norm-based FIR filter has robust performance against disturbances by minimizing the norm of the gain, and in this paper, we propose a numerical method to optimize the Frobenius norm gain. The superior performance of the new DPLL is verified through performance comparison with the existing DPLL through simulation.

그룹 집중 기술로 개선된 Trans-Unet기반 단일 영상 연무제거 신경망

홍찬의(Chan Eui Hong),최현덕(Hyun Duck Choi) 대한전자공학회 2022 전자공학회논문지 Vol.59 No.6

최근 컴퓨터 비전 기술이 발달하면서, 자율주행 분야에서 인공지능 기반 object detection, image segmentation 등의 컴퓨터 비전기술들이 주목받고 있다. 그러나 이러한 기술들은 야간이나 폭우, 안개 등 기후 악조건 주행환경에서는 영상 손실문제로 인해 성능이 급격하게 저하되고, 이는 치명적인 인명피해를 야기시킬 수 있다. 본 논문에서는 이러한 영상 악조건 속에서도 강인한 컴퓨터 비전기술을 확보하기 위해 Group Attention Block (GAB)을 제안하고 이를 Unet 구조와 Vision Transformer에 적용한 새로운 영상 연무제거 모델을 제안한다. 기존의 CNN(Convolution Neural Network)기반 encoder, decoder와 skip connection이 적용된 Unet구조를 통해 영상의 공간적인 정보를 활용한 특징맵을 추출하고 GAB를 적용하여 특징맵을 강화하며, 여기에 Vision Transformer를 추가 적용함으로써 inductive bias를 줄여 글로벌한 정보에서도 영상의 손실이 없도록 개선하였다. 제안하는 신경망 구조는 이전에 연구된 image dehazing 모델에 비해 PSNR(Peak Signal-to-Noise Ratio)과 SSIM(Structural Similarity Index Measure)에서 개선된 결과를 나타냄을 보여준다. With the recent development of computer vision technology, computer vision technologies such as artificial intelligence-based object detection and image segmentation are attracting attention in the field of autonomous driving. However, these technologies degrade performance due to image loss in driving environments under adverse weather conditions such as nighttime, heavy rain, and fog, which can cause fatal human casualties. In this paper, we propose a novel Group Attention Block (GAB) and a haze removal model combined with the Unet and Vision Transformer in order to get robust computer vision technologies even in such adverse image conditions. This network can capture the image feature with spatial information by CNN layer as well as capture global relations without inductive bias through Vision Transformer. Finally, GAB enhances these functions and helps the decoder to restore the clean image. The simulation results show an improvement in dehazing by PSNR (Peak Signal-to-Noise Ratio) and SSIM (Structural Similarity Index Measure) score compared to the previous image dehazing models.

유한 임펄스 응답 필터의 방위각을 이용한 표적추적에 응용

민경민(Kyung-Min Min),최현덕(Hyun-Duck Choi),안춘기(Choon-Ki Ahn),임묘택(Myo-Taeg Lim) 대한전기학회 2013 대한전기학회 학술대회 논문집 Vol.2013 No.10

유한 임펄스 응답 필터의 방위각을 이용한 표적추적에 응용

민경민(Kyung-Min Min),최현덕(Hyun-Duck Choi),안춘기(Choon-Ki Ahn),임묘택(Myo-Taeg Lim) 대한전기학회 2013 정보 및 제어 심포지엄 논문집 Vol.2013 No.10

2계층 Frobenius norm 유한 임펄스 응답 필터 기반 디지털 위상 고정 루프 설계

김신(Sin Kim),신성(Sung Shin),유성현(Sung-Hyun You),최현덕(Hyun-Duck Choi) 한국전자통신학회 2024 한국전자통신학회 논문지 Vol.19 No.1

디지털 위상 고정 루프는 디지털 위상 검출기, 디지털 루프 필터, 디지털 제어 발진기, 분배기 등으로 이루어진 일반적인 회로로 전기 및 회로 분야 등 다양한 분야에서 널리 사용된다. 디지털 위상 고정 루프의 성능 향상을 위해 다양한 수학적인 알고리즘 등을 활용한 상태 추정기가 사용된다. 전통적인 상태 추정기로는 무한 임펄스 응답 상태 추정기의 칼만 필터를 활용해왔으며, 무한 임펄스 응답 상태 추정기 기반 디지털 위상 고정 루프는 초기값의 부정확성, 모델 오차, 다양한 외란 등의 예상치 못한 상황에서 급격한 성능 저하가 발생할 수 있다. 본 논문에서는 새로운 디지털 위상 고정 루프를 설계하기 위해 2계층 Frobenius norm 기반 유한 임펄스 상태 추정기를 제안한다. 제안한 상태 추정기는 첫 번째 층의 추정 상태를 이용하여 두 번째 층에서 상태 추정을 하는데, 이때 첫 번째 층의 추정 상태와 누적된 측정값과 결합하여 설계하였다. 새로운 유한 임펄스 응답 상태 추정기 기반 디지털 위상 동기 루프의 강인한 성능을 검증하기 위해 잡음 공분산 정보가 부정확한 상황에서 무한 임펄스 응답 상태 추정기와 비교하여 시뮬레이션을 수행하였다. The digital phase-locked loop(DPLL) is one of the circuits composed of a digital detector, digital loop filter, voltage- controlled oscillator, and divider as a fundamental circuit, widely used in many fields such as electrical and circuit fields. A state estimator using various mathematical algorithms is used to improve the performance of a digital phase-locked loop. Traditional state estimators have utilized Kalman filters of infinite impulse response state estimators, and digital phase-locked loops based on infinite impulse response state estimators can cause rapid performance degradation in unexpected situations such as inaccuracies in initial values, model errors, and various disturbances. In this paper, we propose a two-layer Frobenius norm-based finite impulse state estimator to design a new digital phase-locked loop. The proposed state estimator uses the estimated state of the first layer to estimate the state of the first layer with the accumulated measurement value. To verify the robust performance of the new finite impulse response state estimator-based digital phase locked-loop, simulations were performed by comparing it with the infinite impulse response state estimator in situations where noise covariance information was inaccurate.

중학생의 식생활 단원에 대한 인식과 식행동에 관한 연구

김경애,최현덕 한국 가정과 교육 학회 1999 한국가정과교육학회지 Vol.11 No.2

The purpose of this study was to investigate the perception and eating behaviors of middle school students on Kwangju area about food and nutrition section of Home Economics course. The subjects were 480 middle school students. The major results were as follows; On the perception of food and nutrition section, the 1st graders learned 'how to settle the problems related to food and nutrition and food choices.' The 2nd grades learned 'the knowledge and principle of food and nutrition for improving dietary life.' The 3rd graders learned 'actual dietary life functions like arranging menu and preparing meals.' Consequently, there was a significant difference ineach grade and all boy and girl students have the highest response to the item 'they learn how to settle the nutrition problem and foods choices.' On the interests in the food and nutrition section, lower graders and girl students have higher interest than higher graders and boy students. The area which was considered as most interesting in food and nutrition sections was 'cooking' in all students. The perception of the concern on dietary life after learning food and nutrition section, the 1st graders responded there was a great change in 'balanced dietary life', the 2nd graders in 'eating habit and nutrition of adolescence' and the 3rd graders in 'cooking'. Especially, girl students showed more changes. On the perception of degree which food and nutrition section gives a help to actual life, students over 50% responded that it was helpful to their actual life, This showed that the lower the grades, the more help it grove to those students. On the perception of mother's job, the students with working mother considered that they learned 'actual dietary life functions like arranging menu and preparing meals' from the food and nutrition section. In addition, they had high interests in the area of 'cooking'. It was recognized that as the family size was greater, the food and nutrition section gave more helps to actual life. On eating behaviors of breakfast, the higher the grads, the more frequent they don't have breakfast, boy students have more frequencies of having breakfast than girl students and the primary reason of not having breakfast is poor appetite. On whether going without a meal affects study or not, most of the 1st, 2nd and 3rd graders whether they are boys and girls or not responded it affected their study. On eating behaviors of lunch, most of boy and girl students in the 1st. 2nd and 3rd grade have lunch box. On eating behaviors of supper. eating time is usually between 7:00 and 8:00 p.m and eating hour is mostly 10 or 15 minutes. On eating behaviors between meals, frequencies of eating between meals and mostly once or twice a day and girl students have more frequencies of eating between meals, and the cause was hunger commonly in boy and gill students and 'killing time' in girl students, The focus in getting the foods is 'taste' and favorite food is fruit regardless of the grade.

근외분광분석법에 의한 옥배유, 팜유, 돈지의 산가, 요오드가, 비누화가 측정

홍진환,최재천,최현철,강명희,최장덕,양창숙,김광수,백성열,김보경,오해성,이희식,조래광 식품의약품안전청 1997 식품의약품안전청 연보 Vol.1 No.-

식용유지의 산가(AV), 비누차가(SV), 요오드가(Iv)를 신속 ·간편하게 출정하기 위해서 1,100nm에서 2,Soonm까지의 근적외 fNIR)툴과반사분광분석법을 응용하였다. 옥배유, 팜유, 돈지를 공기중에서 200·C로 가열산화 처리하여 가열처리하지 않은 유지와 동일종끼리 혼합하여 적정법에 의한 AV,통V,IV의 분석치가 고르게 분포되도록 시료를죠제하였다. 모든 시료는 검량선 작성용 시료군(Calibration set)과 미지시료군(Prediction set)으로 나누었다. Cali-bration ste에 대하여 NIR스펙트럼 데이터와 적정법에 의한 화학분석치 사이에 중회귀분석한 결과, ?거의 검량선이라고 판단되는 중희귀식의 중상관계수는 AV의 경우 0.936,SV는 0.903,IV는 0.979이었다. 이들 검량선으로부터 Prediction set의 AV, SV, IL'를 각각 산출하여 기존 적정법에 의한 화학분썩치와 비교한 결과, 각각의 측정오차는 0.156, 1.713, 1.259이었고, 기존의 적정법과 NIR분광분석법 사이의 상관계쓴(R)은 각각 9.942, 0.880, 0.980이었다. 이와같은 결과로부터 기존꽈 적정법을 대신하여 신속 ·간편하게 측정할 수 있는 NIR분광분석법의 이용가능성이 인정되었다. Near-infrared(NIR) transflectance spectroscopy scanning from 1100nm to 2500nm was applied for the analyses of acid value(AV) , saponification value (SV) and iodine value(IV) in corn oil, palm oil and lard. These samples were allowed to stand at 200℃ in air. Their treated samples mixed with untreated sanples onB the basis of maximum variation in the reference values analyzed by titration method. All samples were divided into calibration set and prediction set. The multiple regression equations(MREs) established for the calibration set between the reference data analyzed by titration method and the NlR spectrl data gave the multiple correlation coefficients(MCCs) of 0.936, 0.903 and 0.979 for AV, SV and IV, respectively. Standard errors of predictions(SEPs) between the data calculated from the best MREs and the reference data analyzed by titration method for the prediction samples were 0.156, 1.713 and 1.289, and the correlation coefficients(R) between those were 0.942, 0.886 and 0.980 for AV, SV and IV, respectively. The obtained results indicate that the NIR transflectance procedure can potentially be used as a nondestructive analysis method for the purpose of rapid and simple measurement of AV, SV and IV is corn oil, palm oil and lard.