http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
주백석(Baeksuk Chu),홍대희(Daehie Hong) 대한기계학회 2011 대한기계학회 춘추학술대회 Vol.2011 No.5
This paper introduces a new trial to apply a robotic system to construction automation. For the purpose, a robotic bolting system is developed to perform beam assembly work of a building. The suggested robot system is composed of a mechanism paπ and a control part. In the mechanism part, a bolting end-effector has the function of assembling steel beams with two types of bolting tools, and a robotic manipulator translates the bolting end-effector to target bolting positions. The control part is based on the visual servoing technology which recognizes the bolting positions with a CCD camera and image processing and sends the position information to the motion controller. The developed robotic bolting system was evaluated with real experiments and showed successful results for beam assembly.
수동 동역학 보행로봇을 위한 강화학습 기반의 제어기 설계
주백석(Baeksuk Chu),홍대희(Daehie Hong),박주영(Jooyoung Park) 대한기계학회 2006 대한기계학회 춘추학술대회 Vol.2006 No.6
Passive dynamic walker is a kind of bipedal walking robots that are able to walk stably down a small decline without using any actuators. The purpose of this research is to design a controller in order to build actuated robots capable of walking on flat terrain based on passive dynamic walking. To achieve the objective, the control algorithm used in this research is reinforcement learning (RL) method. RL is a goal-directed learning of a mapping from situations to actions. The goal of RL is to maximize a reward which is an evaluative feedback from the environment. Constructing the reward of the actuated passive dynamic walker, the control objective, which is stable walking on level ground, is included. In this research, RL algorithm based on actor-critic architecture and gradient-following algorithm is applied. The control algorithm is verified with computer simulations. It is confirmed that with the suggested controller, the control objective is sufficiently achieved.