2022 Fiscal Year Final Research Report
Understanding Proficiency Skills in Handwork Using Motion Constraints
Project/Area Number |
20K11966
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Review Section |
Basic Section 61030:Intelligent informatics-related
|
Research Institution | National Institute of Advanced Industrial Science and Technology |
Principal Investigator |
Yamanobe Natsuki 国立研究開発法人産業技術総合研究所, 情報・人間工学領域, 主任研究員 (90455436)
|
Co-Investigator(Kenkyū-buntansha) |
藤井 浩光 千葉工業大学, 先進工学部, 准教授 (30781215)
|
Project Period (FY) |
2020-04-01 – 2023-03-31
|
Keywords | 技能継承 / 動作解析 / 形式知化 / 動作モデル |
Outline of Final Research Achievements |
This study aims to clarify proficiency skills in handwork and to formalize them in a form that can be preserved and passed on. We conducted skill analysis and modeling of knitting, which requires detailed control of movements in response to reaction forces applied to the fingertips and for which skill evaluation is possible based on the appearance of the knitting pattern and other visual factors. We proposed evaluation values based on the homogeneity of the knit shape and showed that the yarn-pulling motion is important and that the amount, direction, and force of yarn-pulling must be constant based on the analysis of hand motions and visual information. The knitting motion was divided into multiple unit motions and learned by LSTM, enabling automatic segmentation of the measurement data by the motion model. This contributes to the comparison with skilled workers and the facilitation of training.
|
Free Research Field |
ロボット工学
|
Academic Significance and Societal Importance of the Research Achievements |
少子高齢化社会を背景として,労働人口の減少は日本社会に様々な問題を引き起こしている.伝統工芸をはじめとする手工業分野における技術継承の課題もそのうちの一つである.技能的な側面を学ぶためには,見て・倣って・訓練して,という熟練者の技術を実際に学ぶ過程が重要であるが,熟練者の高齢化が進み直接的な技能継承の機会が急速に減少しているのが現状である.本研究は,学習者独自のトレーニングを可能とする熟練技能のアーカイブに貢献するものである.
|