Unified Control Theory of Humanoid Robot by Hybrid Control including Probabilistic State Transition

2020 Fiscal Year Final Research Report

• Project/Area Number: 18K19802
• Research Category: Grant-in-Aid for Challenging Research (Exploratory)
• Allocation Type: Multi-year Fund
• Review Section: Medium-sized Section 61:Human informatics and related fields
• Research Institution: The University of Tokyo
• Principal Investigator: Yamamoto Ko 東京大学, 大学院情報理工学系研究科, 准教授 (20641880)
• Project Period (FY): 2018-06-29 – 2021-03-31
• Keywords: 二足歩行制御 / ハイブリッド制御

Outline of Final Research Achievements

In this research, mathematical frameworks and algorithms for a simplified model of a humanoid robot were developed, including the control design and stabilizable region analysis explicitly considering discontinuous state transitions during walking, and the model predictive control considering the probability of landing impact forces. Moreover, the transformation method from the simplified model to the detailed whole-body dynamics was also proposed. These frameworks and algorithms were validated through dynamics simulations and experiments. In a hydrostatically-driven humanoid robot, stable biped walking motions with a knee-stretched posture, and stable balancing control against unknown disturbances or on an inclining ground.

Free Research Field

ヒューマノイドロボット

Academic Significance and Societal Importance of the Research Achievements

足の踏みかえに伴い構造が離散的に変化する人型ロボットの力学的な特徴を扱うことができる単純化モデルの制御方法、および単純化モデルと詳細全身モデルの間の変換方法を示した。数学・力学に関する理論的枠組みを構築しただけでなく、それらを実験的に検証し、実際のロボットシステムで安定な二足歩行、バランス制御を実現した。これらの成果は、人型ロボットが実環境で安定に歩行できるための基盤技術につながる。