http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
사용자의 시선과 팔 움직임 특징 정보를 이용한 픽앤플레이스 움직임 의도 추정
주다원(Dawon Ju),배준범(Joonbum Bae) 대한전자공학회 2023 대한전자공학회 학술대회 Vol.2023 No.11
Teleoperated robot perform tasks instead of human operator due to physical limitations or places difficult to go. However, because of kinematic inconsistencies between operator and teleoperated robot, the operator has to physical fatigue and affects the operator performance. To overcome this limitation, one solution is to predict the intention of the human operator through operator’s behavior and assist operator before the end of the work. Therefore, in this paper, we propose and LST-Mbased algorithm to predict operator’s Pick-and-Place intention using gaze and arm movement. We train the proposed model using sequential data and learn LSTM based model and real-time motion recognition result is shown.