Development of a Human-like Walking and Running Robot by Fusing Passive Dynamics Mechanism and Deep Learning

2021 Fiscal Year Final Research Report

• Project/Area Number: 19H02109
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Basic Section 20020:Robotics and intelligent system-related
• Research Institution: Nagoya Institute of Technology
• Principal Investigator: SANO AKIHITO 名古屋工業大学, 工学(系)研究科(研究院), 教授 (80196295)
• Co-Investigator(Kenkyū-buntansha): 池俣 吉人 帝京大学, 理工学部, 准教授 (70467356)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 歩行・走行ロボット / 受動的力学機序 / 深層学習 / ヒト歩行・走行 / ロボティクス

Outline of Final Research Achievements

In this study, we developed walking/running robots of approximately 1.4[m] in height that leads to a unified model. As common units, a unique muscle and tendon group that takes into account muscle synergy, and the elastically hanged wobbling mass in trunk which is unparalleled in other robots, are equipped. The robot successfully extracts natural movement that does not resist the environment (gravity) by minimizing activation and controlling the force. In addition, leg motions based on image were generated by deep learning. The simple physical phenomena derived from the passive walking and the hinged-movement have been sublimated to a more human-like locomotor intelligence.

Free Research Field

ロボティクス

Academic Significance and Societal Importance of the Research Achievements

本研究では，継続的に等身大の歩行ロボットと走行ロボットを研究開発しているが，両方一緒に行っているところはほとんどない．身体・環境・学習（神経）の相互作用の中から新たな力学機序の発見を行い，実機実験でこれを実証したことは学術的意義がある．特に，受動的力学機序と深層学習との融合のさらなる可能性が示唆された．よりヒトに近い運動生成は，歩行支援の技術開発などと常にリンクしており社会的意義がある．この視点での研究開発は，国内外においてほとんどなく，後追いではなく真に対抗できるものと考える．