Tomofumi Tsuji1, Sakmongkon Chumkamon1, Chanapol Piyavichayanon2, Ayumu Tominaga3, Ryusuke Fujisawa1, Abbe Mowshowitz4, Eiji Hayashi1
1Department of Mechanical Information Science and Technology, Kyushu Institute
of Technology, 680-4, Kawazu, Iizuka-City, Fukuoka, 820-8502, Japan
2Department of Creative Informatics, Kyushu Institute of Technology, 680-4
Kawazu, Iizuka, Fukuoka 820-8502, Japan
3Department of Creative Engineering Robotics and Mechatronics Course, National
Institute of Technology Kitakyushu Colllege, 5-20-1 Shii, Kokuraminamiku,
Kitakyushu, Fukuoka, 802-0985, Japan
4Department of Computer Science, The City College of New York, 160 Convent
Avenue, New York, NY 10031, USA
Pages 8–13
ABSTRACT
There is a growing demand for automation by robots in the home replacement
meal industry due to labor shortages in food factories and from the perspective
of the SDGs[1]. In this research, we are developing autonomous work robots
that can perform home replacement meal tasks and developing the technology
for industrial food automation using Artificial Intelligence (AI) to improve
productivity, security, and safety. In this paper, we perform weight estimation
of the served object to identify the amount of spaghetti grasped by the
robot. We created our dataset of spaghetti used for weight estimation.
Spaghetti of varying weight is in different types of containers, placed
at random positions in the robot workspace. The proposed model is shown
to estimate the weight of spaghetti with an error of at most 10%.t
Keywords: FA robots, Image processing, Moodle products, Weight estimation,
Deep learning, General dataset
ARTICLE INFO
Article History
Received 8 November 2021
Accepted 27 June 2022
J-STAGE3102