http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
얇은막대 배치작업에 대한 N-R 과 EKF 방법을 이용하여 개발한 로봇비젼 제어알고리즘의 평가
손재경(Jae Kyung Son),장완식(Wan Shik Jang),홍성문(Sung Mun Hong) 대한기계학회 2013 大韓機械學會論文集A Vol.37 No.4
실제 산업현장에서 비젼 시스템을 적용하기에는 로봇 비젼 제어알고리즘의 기구학모델의 정확도, 로봇이 움직이는 동안 카메라 초점거리와 방위에 대한 보정, 3 차원 물리적 좌표에서 2 차원 카메라 좌표로의 매핑에 대한 이해 등 해결해야 할 많은 문제점들이 있다. 본 논문에 제안된 비젼 시스템 모델은 카메라와 로봇 사이의 상대적인 위치가 알려지지 않아도 제어가 가능하고, 카메라 보정 문제를 해결하기 위해 6 개의 카메라 매개변수를 가지는 비젼 시스템 모델을 제시하였으며, 이를 이용하여 로봇 비젼 제어알고리즘 개발에 N-R 방법과 EKF 방법을 적용하였다. 최종적으로 N-R 과 EKF 방법에 의하여 개발된 로봇 비젼 제어 알고리즘의 위치 정밀도와 데이터 처리 시간을 얇은 막대 배치작업을 수행하여 비교하였다. Many problems need to be solved before vision systems can actually be applied in industry, such as the precision of the kinematics model of the robot control algorithm based on visual information, active compensation of the camera’s focal length and orientation during the movement of the robot, and understanding the mapping of the physical 3-D space into 2-D camera coordinates. An algorithm is proposed to enable robot to move actively even if the relative positions between the camera and the robot is unknown. To solve the correction problem, this study proposes vision system model with six camera parameters. To develop the robot vision control algorithm, the N-R and EKF methods are applied to the vision system model. Finally, the position accuracy and processing time of the two algorithms developed based based on the EKF and the N-R methods are compared experimentally by making the robot perform slender bar placement task.
강체 고정 목표물에 대해 Newton-Raphson알고리즘을 이용한 로봇 비젼제어기법 개발
손재경(Jae Kyung Son),장완식(Wan Shik Jang),홍성문(Sung Moon Hong) 대한기계학회 2012 대한기계학회 춘추학술대회 Vol.2012 No.11
This study is to develop the vision control scheme based on Newton-Raphson(N-R) algorithm and is concerned with the application of the vision control scheme for robots fixed rigid body placement task. This proposed scheme is to make the robot move actively, even if relative position between camera and robot are unknown. This scheme consists of vision system model, camera parameters estimation scheme and robot’s joint-angle estimation scheme. also, the vision system model includes six camera internal and external parameters. The N-R algorithm is based on iterative technique and can concurrently handle large amounts of data. This study uses the four cues for the rigid body placement task to consider the location and orientation. Finally, practicality of robot vision control scheme based on N-R algorithm is verified experimentally by performing the robots position control for the fixed rigid body targets.
손재경(Jae Kyung Son),이동섭(Tong Sup Lee),양한섭(Han Seob Yang) 한국수산과학회 1999 한국수산과학회지 Vol.32 No.5
To understand the role of shelf sediment in phosphorus biogeochemical cycle, we carried out sequential sediment extraction (SEDEX) of P and porewater analysis on 14 core samples collected in the South Sea of Korea. SEDEX classified P-pools into 5 phases and results are grouped into two cateories: reactive P (loosely sorbed-P and Fe bound-P) and refractory P (detrital inorganic-P, authigenic mineral-P and organic-P). Total P concentrations are decreased with sediment depth in all samples as a result of dissolution to porewater. Reactive P comprises about 20∼50% of total P, and iron bound-P is the major form consisting 70∼80% of reactive P-pool. Iron bound-P decreases sharply with depth. Depth profiles of dissolved P concentration in porewater show mirror image of iron bound-P, revealing the role of FeOOH as a regulator of reactive P supply to overlying water culumn. Authigenic mineral-P consists less than 5% of total P, thus removal of reactive P by converting into refractory P seems inefficient in shelf sediment. This implies that continental shelf sediment sequesters P temporarily rather than permanently. Results show local variation. Nakdong estuary receiving large amount of terrigenous input shows the highest concentration of total P and reactive P. Here iron oxyhydroxides at the surface sediment control the water column flux of P from sediment. Although total P content at the surface is comparable (500∼600 ㎍ g^(-1)) between the South Sea and East China Sea, the former contains more iron bound-P and less derital inorganic-P than the latter. Reasons for the difference seem due in part to particle texture, and to biological productivity which depends roughly on the distance from land.
김재명(Jae Myung Kim),손재경(Jae Kyung Son),장완식(Wan Shik Jang) Korean Society for Precision Engineering 2018 한국정밀공학회지 Vol.35 No.7
This paper proposes a review of the robot’s vision control scheme using the data moving method, to improve the accuracy and processing speed for the tracking of a moving target. The vision system model, which can actively adjust the camera parameters for the camera and robot position changes is used for this study. The method of processing the vision data obtained during robot movement toward the target can be classified into a batch method using all the acquired data, and a data moving method using only limited data. In an effort to reduce the number of vision data obtained while the robot moves toward the target, the proposed control scheme estimates the optimal number of robot moving points near the target, to exclude the old data and use only the limited data obtained near the target recently. In this study, limited data are utilized. In order to show the effectiveness of the proposed control scheme, we set the control method based on the batch processing method, and then compared these two results with the accuracy and the processing time by performing the experiments of the slender-bar moving target tracking method.