• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1995년도 추계학술대회 논문집
• /
• pp.1114-1119
• /
• 1995

This paper presents the development of estimation model and control method based on the new computer vision. This proposed control method is accomplished using a sequential estimation scheme that permits placement of the rigid body in each of the two-dimensional image planes of monitoring cameras. Estimation model with six parameters is developed based on a model that generalizes known 4-axis scara robot kinematics to accommodate unknown relative camera position and orientation, etc. Based on the estimated parameters,depending on each camers the joint angle of robot is estimated by the iteration method. The method is tested experimentally in two ways, the estimation model test and a three-dimensional rigid body placement task. Three results show that control scheme used is precise and robust. This feature can open the door to a range of application of multi-axis robot such as assembly and welding.

• Journal of Power Electronics
• /
• 제13권5호
• /
• pp.814-828
• /
• 2013

The LCL resonant inductive power transfer (IPT) system is increasingly used because of its harmonic filtering capabilities, high efficiency at light load, and unity power factor feature. However, the modeling and controller design of this system become extremely difficult because of parameter uncertainty, high-order property, and switching nonlinear property. This paper proposes a frequency and load uncertainty modeling method for the LCL resonant IPT system. By using the linear fractional transformation method, we detach the uncertain part from the system model. A robust control structure with weighting functions is introduced, and a control method using structured singular values is used to enhance the system performance of perturbation rejection and reference tracking. Analysis of the controller performance is provided. The simulation and experimental results verify the robust control method and analysis results. The control method not only guarantees system stability but also improves performance under perturbation.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1997년도 춘계학술대회 논문집
• /
• pp.729-733
• /
• 1997

Due to the variety of disturbance, it is not ease to accomplish the in-process detection of weld line with non-contact sensor. To get around this difficulties problem develop an automatic seam tracking weld system, the reliable signal processing algorithm has been recommanded. In this research, laser displacement sensor is applied as a seam finder in the automatic tracking system. The sensor is controlled by a dc servo motor which is mounted at X-Y moving table. X-Y moving table manipulated by an ac servo motor controls the position and velocity of the welding torch. First, X-Y table moves to Y-axis to search the welding joint feature before starting the welding, and welding joint is from the scanning data and weighting factor for each other. Second, weld line is determined using proposed signal processing algorithm during welding process. Form the experimental results, we could see the possibility that laser displacement sensor with procesed algorithm can be used as a seam finder in welding process under the severe noise (spatter,arc light etc.) condition

• 전자공학회논문지S
• /
• 제34S권4호
• /
• pp.57-65
• /
• 1997

The purpose of this paper is to propose the HMM (hidden markov model) based on multi-observation sequence for the isolated word recognition. The proosed model generates the codebook of MSVQ by dividing each word into several sections followed by dividing training data into several sections. Then, we are to obtain the sequential value of multi-observation per each section by weighting the vectors of distance form lower values to higher ones. Thereafter, this the sequential with high probability value while in recognition. 146 DDD area names are selected as the vocabularies for the target recognition, and 10LPC cepstrum coefficients are used as the feature parameters. Besides the speech recognition experiments by way of the proposed model, for the comparison with it, the experiments by DP, MSVQ, and genral HMM are made with the same data under the same condition. The experiment results have shown that HMM based on multi-observation sequence proposed in this paper is proved superior to any other methods such as the ones using DP, MSVQ and general HMM models in recognition rate and time.

• 한국멀티미디어학회논문지
• /
• 제17권12호
• /
• pp.1418-1429
• /
• 2014

In this paper we suggest a new two-wheelers detection algorithm using local cell features. The first, we propose new feature vector matrix extraction algorithm using the correlation two cells based on local cell histogram and shifting from the result of histogram of oriented gradients(HOG). The second, we applied new weighting values which are calculated by the modified histogram intersection showing the similarity of two cells. This paper applied the Adaboost algorithm to make a strong classification from weak classification. In this experiment, we can get the result that the detection rate of the proposed method is higher than that of the traditional method.

• Journal of Advanced Marine Engineering and Technology
• /
• 제23권5호
• /
• pp.630-636
• /
• 1999

In this paper in optimal modulation controller for position control and anti-sway of container crane systems is designed by a recursive algorithm that determines the state weighting matrix Q of a linear quadratic performance. The optimal modulation controller is based on optimal control. The basic feature of the recursive algorithm is the reduction of the number of iterations as well as minimization of the calculations involved So in order to obtain a mathematical model which rep-resents the equation of motion of the trolley and load Lagrange equation is used. The optimal modulation controller has been verified and simulated to show that it is robust when a load dis-turbance is applied and a reference is changed.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2002년도 추계종합학술대회
• /
• pp.296-300
• /
• 2002

본 연구에서 는 비지도 학습 방식인 SOFM(Self Organize Feature Maps)과 지도 학습인 다층 신경회로망을 이용하여 패턴 분류를 하는 방식을 제안하였다. SOFM을 이용하여 입력 패턴을 분류하여 얻은 결과를 다층 신경회로망의 초기 연결강도와 목표 값으로 설정한다. 제안한 방식의 유용성을 확인하기 위하여 얼굴 영상에 대하여 시뮬레이션한 결과 우수한 성능을 얻었다.

• 한국정밀공학회지
• /
• 제15권11호
• /
• pp.100-107
• /
• 1998

This paper presents the development of estimation model and control method based on the new robot vision. This proposed control method is accomplished using the sequential estimation scheme that permits placement of the rigid body in each of the two-dimensional image planes of monitoring cameras. Estimation model with six parameters is developed based on the model that generalizes known 4-axis scara robot kinematics to accommodate unknown relative camera position and orientation, etc. Based on the estimated parameters, depending on each camera the joint angle of robot is estimated by the iteration method. The method is experimentally tested in two ways, the estimation model test and a three-dimensional rigid body placement task. Three results show that control scheme used is precise and robust. This feature can open the door to a range of application of multi-axis robot such as assembly and welding.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2012년도 추계학술발표대회
• /
• pp.417-420
• /
• 2012

기계학습을 이용하는 문서 자동분류 시스템은 분류모델의 구성을 위해서 단어를 특징으로 사용한다. 자동분류 시스템의 성능을 높이기 위해 보다 의미있는 특징을 선택하여 분류모델을 구성하기 위한 여러 연구가 진행되고 있다. 특히 인터넷상에서 사용되는 웹문서는 단어 외에도 태그정보, 링크정보를 가지고 있다. 본 논문에서는 이 두 가지 정보를 이용하여 웹문서 자동분류 시스템의 성능을 향상 시키는 방법 제안 한다. 태그 정보와 링크 정보를 이용하여 적절한 특징을 선택하고, 각 특징의 중요도를 계산하여 가중치를 구한다. 계산된 가중치를 각 특징에 부여하여 분류 모델을 구성하고 나이브 베이지안 분류기를 통하여 성능을 평가하였다

• 한국컴퓨터정보학회논문지
• /
• 제11권4호
• /
• pp.1-7
• /
• 2006

최근 이미지 검색은 검색의 정확성을 높이고자 사용자의 요구를 반영하는 적합성 피드백에 관한 연구가 활발히 진행되고 있다. 본 논문은 이미지 검색 시 나타나는 고수준 개념과 저수준 특징 사이의 의미적 격차를 줄이기 위하여 적합성 피드백에 기반한 영역 기반 이미지 검색의 가중치 기법에 대해서 논의하고 새로운 가중치 기법을 제안한다. 새롭게 제시된 가중치 기법은 한 이미지에 존재하는 영역들의 공간적 위치에 따라 영역의 중요성을 결정한다. 실험 결과는 본 논문에서 제시된 가중치 기법이 평균 재현율에 있어서 크기 백분율 가중치 기법에 비해 약 18%, 역 이미지 빈도수를 적용한 영역 빈도수 가중치 기법에 비해 약 11% 가량 높게 나타나는 것을 보이고 있으며, 검색 시간에 있어서도 영역 빈도수 가중치에 비해 약 1/10인 것을 보이고 있다.