http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Ko, Eunseok,Hong, Daesik IEEE 2007 IEEE Transactions on Vehicular Technology VT Vol.56 No.2
<P>In this paper, we propose a robust space-time block coded (STBC)-based transmit diversity scheme with low decoding complexity against spatially transmit-correlated fading channels compared with the conventional STBC orthogonal frequency-division multiplexing (STBC-OFDM) schemes. The proposed scheme is comprised of Alamouti code for spatial diversity and a simple mapping process with subcarrier selection for frequency diversity. Proceeding from a detailed analysis of the rank and determinant of the channel correlation matrix, we present a proper subcarrier selection scheme satisfying the conditions for robustness against spatial transmit correlation. Furthermore, when the receiver employs a low-complexity decoder such as a zero-forcing decoder, the proposed scheme shows improved performance compared with the existing space-frequency coded OFDM schemes</P>
Understeer 상황에서 차량 안정성 향상을 위한 Brake Torque Vectoring에 대한 연구
고은석(Eunseok Ko),이강원(Kangwon Lee),김형수(hyoungsoo Kim) 한국자동차공학회 2017 한국자동차공학회 학술대회 및 전시회 Vol.2017 No.11
This paper study Brake Torque Vectoring that aims at enhancing vehicle stability and dynamic behavior in understeer condition. Recently, the vehicle makers in market introduce brake base torque vectoring function as a main contributor to vehicle dynamic management to improve cornering performance and agility from normal to critical driving. Brake Torque Vectoring is a function that can apply to a vehicle with low development cost and a short period of development time. The function can carry out a similar performance to Differential (Mechanical) Torque Vectoring System for highest performance application without additional hardware. This study will present improved agility and better cornering stability by comparing dynamic behavior through a development vehicle with optimally tuned Brake Torque Vectoring Function.
차량 모델 개선을 위한 Wheel Alignment 영향에 관한 연구
고은석(Eunseok Ko),이강원(Kangwon Lee),김형수(Hyoungsoo Kim),오승규(Seungkyu Oh) 한국자동차공학회 2014 한국자동차공학회 부문종합 학술대회 Vol.2014 No.5
Vehicles with Electronic Stability Control (ESC) sold in ECE countries must be evaluated with ECE Regulation 13H, which specifically encourages the use of simulation. Because a position of simulation increase as above, it required validated simulation vehicle model with performance same as actual vehicle. However, due to differences between actual vehicle and simulation model, the industry has done a lot of work to correlate these two factors. They are trying to working for correlation with actual vehicle performance. So. This paper is aimed to figure out how wheel alignment is important factor for reducing the difference between simulation and physical test by analyzing its effect. The value of Wheel alignment is calculated by DOE Kriging Method for analyzing influence of control factor and by optimization to minimize difference with physical test.
HIL Simulator를 이용한 BAS(Brake Assist System) 평가방법 연구
고은석(Eunseok Ko),오승규(Seungkyu Oh),김형수(Hyungsoo Kim) 한국자동차공학회 2015 한국자동차공학회 부문종합 학술대회 Vol.2015 No.5
According to the analysis of the rear collision, it showed that the accident was related by a lack of brake pedal effort when braking. Therefore, BAS(Brake Assist System) was developed to supplement or lack of a driver`s braking pedal pressure. Especially, In Europe (ECE R13H) and Korea (KMVSS 90) regulation, it was mandatory to apply BAS System to all of the vehicle which has been produced from 1 January 2015. In this reason, the industry has done a lot of work to develop and evaluate the BAS system. Especially in physical vehicle, it is difficult to evaluate the vehicle under same conditions, because, there are a lot of disturbance like driver`s ability and ware condition of tire. This study suggested the method for evaluating the BAS system using HIL Simulator that is possible to conduct test under always same conditions and analyzed the parameter affecting the results.
Action Recognition Model to Monitor Illegal Dumping using Zoom-In Image Sequence Data
Kyoungmin Ko,Hyunmin Gwak,Eunseok Lee,Gunhwi Kim,Donghyeon Moon,Youngjoo Cho,SungHwan Kim 계명대학교 자연과학연구소 2021 Quantitative Bio-Science Vol.40 No.2
In this study, we propose an action recognition model that provides generalized performance regardless of camera location and distance between the camera and human. The proposed model consists of two-stage networks, namely, human detection and action recognition. The proposed method operates on video frames that are resized by a new zoom-in method using pretrained Yolo v3. To use temporal information, which is regarded as a critical factor in action recognition, we adopt the R(2+1)D model, which is a factorized model capable of representing more complex networks. The proposed Zoom-In method yields generalized performance regardless of distance. In an experiment, the proposed method exhibited accuracies of 96.07%, 96.61%, and 94.55% in the short, medium, and long ranges in which our datasets were employed, respectively.