Motor learning based on safe interactions between a human and a soft robot

2023 Fiscal Year Final Research Report

• Project/Area Number: 19KK0285
• Research Category: Fund for the Promotion of Joint International Research (Fostering Joint International Research (A))
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 20020:Robotics and intelligent system-related
• Research Institution: Kyushu Institute of Technology
• Principal Investigator: Ikemoto Shuhei 九州工業大学, 大学院生命体工学研究科, 准教授 (00588353)
• Project Period (FY): 2020 – 2023
• Keywords: ソフトロボット / 運動学習 / テンセグリティロボット

Outline of Final Research Achievements

In this research project, we have developed a soft and durable robot platform and tackled the motor learning in a human-robot interaction scenario. Due to the COVID-19 disaster that occurred shortly after the start of the project, we gave up on developing the platform and conducting experiments at the collaboration partner's laboratory, and changed our policy to conducting the experiments at the organization to which the principal investigator belongs (Kyushu Institute of Technology). The majority of the research was conducted remotely, but a short-term trip was conducted in 2022, and through close discussions and intensive research over the visiting term, several academic results were successfully achieved. We have reported on kinematics learning exploiting the advantages of the robot platform as multiple conference proceedings.

Free Research Field

ロボティクス

Academic Significance and Societal Importance of the Research Achievements

ソフトロボットは，ロボット自体をやわらかさによって人やモノに安全に触れることができるが，一般に制御が難しい．本研究では，やわらかく耐久性に優れたロボットプラットフォームとしてテンセグリティマニピュレータを開発し，その特徴を活かして問題に対処する方法として，長時間ランダムに動いた経験，人やモノに触れ合った経験から，制御系を学習によって獲得する枠組みに注目した．その結果，テンセグリティマニピュレータの設計・実装に関する新しい知見や，その手先の位置や体の剛性，状態の推定など，多くの機能を実際に学習によって獲得できることなどが明らかになった．