This study explores how to develop a digital twin model for factory automation modules and a two-arm collaborative robot and how to utilize the model.&nbsp;Factory automation modules used in this study are a training system for Industry 4.0, which&nbsp;includes SIF-401 (pallet and container feeding station), SIF-402 (container filling station –&nbsp;solid), SIF-403 (container filling station – liquid), SIF-404 (container filling station –&nbsp;customized product), SIF-405 (capping station), SIF-406 (container warehouse station), and&nbsp;SIF-407 (container labeling and dispatching station) by SMC®. These modules were used to&nbsp;study how to construct a digital twin and what kind of information is needed to describe the&nbsp;realistic behavior of the digital model of the factory automation modules.The team focused on the factory automation modules for digital twin modeling at the beginning of this study but realized that the usage of CAD drawings was limited and the software support was also limited.&nbsp;In addition, the team realized that the digital twin of the factory automation modules can be used for virtual commissioning but there were not many things the team could do using the digital twin model. Then, the team switched the focus to a two-arm collaborative robot. The robot used in this study is IRB 14000 Yumi by ABB.&nbsp;Unlike most other robots, Yumi has two arms each of which has seven links and a gripper. Yumi is designed to work in collaboration with a human&nbsp;rather than fully automate all the work. A human can teach Yumi a task by manually operating the robot arms. Yumi can mimic what human arms can do and it opens up a lot of possibilities. The digital twin model of Yumi can be used for virtual commissioning. Various experimental tasks can be tested virtually without taking the risk of causing the robot malfunction. The digital twin model can find the optimal sequence of motions to fulfill a task using deep learning.This study extended the digital twin model of Yumi to use human motion as an input. After reading the coordinates and their&nbsp;changes of human arms over time, the collected data should be transferred to the digital twin of Yumi. One challenge is human arm motions to robot arm motions since human arms and robot arms have different configurations. Human arms have fewer links but the joints are more flexible than those of robot arms. One way to do it is using deep learning and this study is exploring ways to implement the idea.

Human Motion to Collaborative Two-Arm Robot Through Digital Twin Models

Paper Proceedings: https://asmedigitalcollection.asme.org/IMECE/IMECE2021/volume/85543

ASME 2021 International Mechanical Engineering Congress and Exposition (IMECE2021)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.