http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
류제두,남건석,하경남 한국산업융합학회 2020 한국산업융합학회 논문집 Vol.23 No.3
Recently, various tries to apply ROV (Remotely Operated Vehicle) into underwater are being developed. However, due to underwater environment uniqueness, the additional problem must be taken into account when designing an ROV for the inspection of the underwater structure. This is because a GPS-based information method cannot be applied, and the obtainable image is also dependent on the turbidity. Also, it is necessary to be able to satisfy waterproof and operating speeds in consideration of most practical application environments. This paper describes the design results of the ROV system for underwater structure inspection considering the above problems. The designed system applied INS / DVL for location recognition and was configured to support 3D mapping and stereo camera-based image information using sonar depending on visibility. To satisfy the waterproof, a pressure vessel using a composite material was applied. And over-actuated system using eight thrusters to maintain a stable posture and operating speed was applied also. The designed system was verified by structural analysis and flow analysis also.
LMI 를 이용한 2 축 평면 기구 로봇의 ILC 제어기 설계
류제두,윤종현,박종현 제어로봇시스템학회 2013 제어로봇시스템학회 국내학술대회 논문집 Vol.2013 No.5
In general, industrial robots carry out repetitive tasks on production lines. Although robot control technology has really improved, tracking error still occurs when a robot performs repetitive tasks under the influence of iterative noise and other disturbances. To solve these problems, Iterative Learning Control (ILC) was proposed. ILC is a learning control that can reduce the error of the current motion by using the error of the previous run. ILC using Linear Matrix Inequality (LMI), the one of ILC algorithms, can be applied to designing a simple controller in the frequency domain of complex systems. In this paper, learning function (L-function) of ILC is designed by using an LMI algorithm and control performance was compared for different various orders of L-function. Also, Q-filter is designed based on the dynamics of the designed system. To confirm the improved performance of the algorithms, designed ILC controller is applied to a 2-link planer robot manipulator in simulations.