http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
자율주행의 기능 안전을 위한 슬라이딩 모드 관측기 기반 가중 가속도 고장 탐지 및 허용 제어 알고리즘
송태준(Taejun Song),오광석(Kwangseok Oh),이종민(Jongmin Lee),윤영민(Youngmin Yoon),이경수(Kyongsu Yi) 대한기계학회 2018 대한기계학회 춘추학술대회 Vol.2018 No.12
This paper describes a weighted acceleration fault detection and tolerant control algorithm for functional safety of autonomous driving based on sliding mode observer. In order to detect fault signal of longitudinal acceleration, fault detection regions have been reconstructed using multi-sliding mode observer. The weighted fault diagnosis method has been proposed in this study to enhance diagnosis reliability. The fault signal in acceleration has been reconstructed based on sliding mode observer using information of the preceding vehicle acceleration. Based on the reconstructed fault signal, fault-tolerant control algorithm has been constructed for functional safety of autonomous driving. The 3D vehicle model and actual driving data have been used to conduct reasonable performance evaluation.
4 휠 독립 조향 차량의 핸들링 성능 향상을 위한 자가 동조 기반 적응형 모델 예측 제어 알고리즘 개발
송태준(Taejun Song),이혜원(Hyewon Lee),김민준(Minjun Kim),오광석(Kwangseok Oh) 한국자동차공학회 2020 한국 자동차공학회논문집 Vol.28 No.8
This paper presents a self-tuning-based adaptive model predictive control algorithm for improving the handling performance of four-wheel independent- steering vehicles. The model predictive control algorithm was adopted to independently compute the optimal four-wheel steering angles of a vehicle for its handling performance by considering the physical constraints, such as the steering limits and the opposite-direction steering condition. The vehicle nonlinear model was used, and its Jacobian matrices were utilized to compute the control input for determining of the independent steering wheel angle with the driver’s steering input. Due to the model uncertainty and calculation process of the model predictive control algorithm, the uncertainty is increasing in the prediction step of the control algorithm, which may result in a relatively low control performance. To overcome the limitation by model uncertainty of the model predictive control algorithm, self-tuning and adaptation algorithms were designed based on the recursive least squares and MIT rule. Performance evaluation was conducted under a step steering scenario with various velocity conditions.
Structural Reliability of a Suspension Bridge Affected by Environmentally Assisted Cracking
Taejun Cho,Myung-Kwan Song 대한토목학회 2006 KSCE journal of civil engineering Vol.10 No.1
This paper deals with the environmentally assisted cracking behavior of high strength cable wires of a suspension bridge, exposed to marine environments. The main objective is to determine the safety of a bridge system locally and globally. To evaluate the safety of a suspension bridge, an ultimate limit state for the local cracking in cable wires and service limit states for global responses of the bridge system are considered. In an ultimate limit state, the reliability of time-dependent and crack length-dependent Environmentally Assisted Cracking (EAC) of a cable wire has been calculated. Global system responses have been evaluated by a FEM program, which were used for the safety evaluation of service limit states. Due to the diffusion of hydrogen atoms affected by the gradient of hydrostatic stress, the section of cable wire will fail immediately when the ratio of EAC in a cable wire is larger than 60percent. However, the global responses show ignorable differences in the considered analysis. Suggestions for the found problems are discussed.