http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Yee Yeng Liau(료이잉),Kwangyeol Ryu(류광열) 대한산업공학회 2021 대한산업공학회 춘계학술대회논문집 Vol.2021 No.6
Human-robot collaboration (HRC) systems have become one of the enabling technologies in Industry 4.0. With the development of deep learning algorithms, researchers implement various recognition systems into the HRC systems to enable the interaction between humans and robots in HRC work cells, such as object recognition to identify parts and tools, gesture recognition to identify commands using gesture. This paper proposes a conceptual framework of task recognition and progress estimation using objects and hand motion recognition for HRC in mold assembly operations. Task recognition consists of object recognition on parts and tools to identify tasks. With task recognition, progress estimation predicts the status of the task being executed by tracking the recognized parts, tools, and motion. A human controls tasks of robots in an HRC assembly cell using a push button in most practical cases. However, this kind of control method causes delays during the operation. We can eliminate the delays by executing robot tasks including picking and feeding parts before the manual task ends. Therefore, we integrate progress estimation into task recognition to reduce delays during the collaborative operation by deciding the start time of the subsequent robot tasks based on the status of the manual task and vice versa. The outcome of this framework will be helpful to enable just-in-time task execution without commands from a human operator.