• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Unified Understanding through Analysis and Verification of Human Environmental Adaptation Learning Methods

Research Project

Project/Area Number 20KK0256
Research Category

Fund for the Promotion of Joint International Research (Fostering Joint International Research (A))

Allocation TypeMulti-year Fund
Review Section Basic Section 20020:Robotics and intelligent system-related
Research InstitutionTohoku University

Principal Investigator

Hayashibe Mitsuhiro  東北大学, 工学研究科, 教授 (40338934)

Project Period (FY) 2021 – 2023
Project Status Completed (Fiscal Year 2023)
Budget Amount *help
¥11,700,000 (Direct Cost: ¥9,000,000、Indirect Cost: ¥2,700,000)
Keywords運動学習 / 模倣学習 / 深層強化学習 / 同期現象 / 運動シナジー / CPG / 環境適応 / 運動制御
Outline of Research at the Start

本研究の基課題は、環境適応可能なリズミックな運動生成をどのような計算方法で実現できるかという環境協調運動制御メカニズムを解明するものである。環境情報が未知でも、人間が関節間の協調運動を学習アルゴリズムで発現することができる手法を開発している。四足歩行動作ではその運動学習法の有効性を示すことができたが、本国際共同研究ではさらに一歩進んで、一般的な身体モデルの同期現象の学習による生成モデルへと拡張し、多自由度の同期現象を統一的に理解する数理モデルを構築する。またヒトの運動学習特性を再現できるかどうかも検証する必要性があり、その実現に向け国際共同研究を行う。

Outline of Final Research Achievements

The main theme of this research, "Mechanism of Expression of Diverse Movements through Environment-Adaptive Learning Independent of Environment Models and Oscillator Models," aims to elucidate the mechanism of environment-coordinated motion control by investigating how environment-adaptive rhythmic motion generation can be achieved through what computational methods. This international collaborative research has yielded results contributing to a unified understanding of human environmental adaptation learning methods from three perspectives. With Professor d'Avella from Italy, achievements have been made in understanding and replicating the synergy structure of whole-body movement expression processes. With Professor Burdet from the UK, achievements have been made in enhancing motor learning efficiency through shared tactile information. With Professor Ijspeert from Switzerland, foundational technologies for developing AI capable of mimicking biological movements have been developed.

Academic Significance and Societal Importance of the Research Achievements

近年、深層強化学習や模倣学習をそれぞれ用いたロボット制御の応用研究が活発に行われ注目されています。深層強化学習を活用する場合には環境適応可能な運動が生成できるものの、広大な入力空間の探索に膨大な計算コストを要することが問題となります。一方、模倣学習を用いる場合には学習した運動に近い範囲に環境適応性が制限されるという問題が一般的に知られています。今回の提案手法は深層強化学習と模倣学習の両面の利点を生かすことができ、またその欠点を補いあうことができる新しい運動生成の手法となり、多自由度系で生体の自己組織的な振る舞いの生成をAIにより実装する技術につながります。

Report

(4 results)
  • 2023 Annual Research Report   Final Research Report ( PDF )
  • 2022 Research-status Report
  • 2021 Research-status Report
  • Research Products

    (30 results)

All 2024 2023 2022 2021 Other

All Int'l Joint Research (4 results) Journal Article (10 results) (of which Int'l Joint Research: 2 results,  Peer Reviewed: 10 results,  Open Access: 8 results) Presentation (14 results) (of which Int'l Joint Research: 7 results) Remarks (2 results)

  • [Int'l Joint Research] Imperial College London(英国)2024

    • Year and Date
      2024-03-07
    • Related Report
      2023 Annual Research Report
  • [Int'l Joint Research] Fondazione Santa Lucia(イタリア)2023

    • Year and Date
      2023-06-04
    • Related Report
      2023 Annual Research Report
  • [Int'l Joint Research] Imperial College London(英国)2023

    • Year and Date
      2023-05-28
    • Related Report
      2023 Annual Research Report
  • [Int'l Joint Research] Imperial College London(英国)2022

    • Year and Date
      2022-04-09
    • Related Report
      2023 Annual Research Report
  • [Journal Article] Speed-Variable Gait Phase Estimation During Ambulation via Temporal Convolutional Network2024

    • Author(s)
      Guo Yan、Hutabarat Yonatan、Owaki Dai、Hayashibe Mitsuhiro
    • Journal Title

      IEEE Sensors Journal

      Volume: 24 Issue: 4 Pages: 5224-5236

    • DOI

      10.1109/jsen.2023.3343721

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Integrated Quantitative Evaluation of Spatial Cognition and Motor Function with HoloLens Mixed Reality2024

    • Author(s)
      Tada Kenya、Sorimachi Yuhei、Kutsuzawa Kyo、Owaki Dai、Hayashibe Mitsuhiro
    • Journal Title

      Sensors

      Volume: 24 Issue: 2 Pages: 528-528

    • DOI

      10.3390/s24020528

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] AI-CPG: Adaptive Imitated Central Pattern Generators for Bipedal Locomotion Learned through Reinforced Reflex Neural Networks2024

    • Author(s)
      Guanda Li, Auke Ijspeert, Mitsuhiro Hayashibe
    • Journal Title

      IEEE Robotics and Automation Letters

      Volume: -

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Soft-body dynamics induces energy efficiency in undulatory swimming: A deep learning study2023

    • Author(s)
      Li Guanda、Shintake Jun、Hayashibe Mitsuhiro
    • Journal Title

      Frontiers in Robotics and AI

      Volume: 10

    • DOI

      10.3389/frobt.2023.1102854

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Synergy-Space Recurrent Neural Network for Transferable Forearm Motion Prediction from Residual Limb Motion2023

    • Author(s)
      Ahmed Muhammad Hannan、Chai Jiazheng、Shimoda Shingo、Hayashibe Mitsuhiro
    • Journal Title

      Sensors

      Volume: 23 Issue: 9 Pages: 4188-4188

    • DOI

      10.3390/s23094188

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Transhumeral Arm Reaching Motion Prediction through Deep Reinforcement Learning-Based Synthetic Motion Cloning2023

    • Author(s)
      Ahmed Muhammad Hannan、Kutsuzawa Kyo、Hayashibe Mitsuhiro
    • Journal Title

      Biomimetics

      Volume: 8 Issue: 4 Pages: 367-367

    • DOI

      10.3390/biomimetics8040367

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Autonomous Navigation System in Pedestrian Scenarios Using a Dreamer-Based Motion Planner2023

    • Author(s)
      Zhu Wei、Hayashibe Mitsuhiro
    • Journal Title

      IEEE Robotics and Automation Letters

      Volume: 8 Issue: 6 Pages: 3836-3843

    • DOI

      10.1109/lra.2023.3273514

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Self-organizing neural network for reproducing human postural mode alternation through deep reinforcement learning2023

    • Author(s)
      Shen Keli、Li Guanda、Chemori Ahmed、Hayashibe Mitsuhiro
    • Journal Title

      Scientific Reports

      Volume: 13 Issue: 1

    • DOI

      10.1038/s41598-023-35886-y

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] A Hierarchical Deep Reinforcement Learning Framework With High Efficiency and Generalization for Fast and Safe Navigation2023

    • Author(s)
      Zhu Wei、Hayashibe Mitsuhiro
    • Journal Title

      IEEE Transactions on Industrial Electronics

      Volume: 70 Issue: 5 Pages: 4962-4971

    • DOI

      10.1109/tie.2022.3190850

    • Related Report
      2022 Research-status Report
    • Peer Reviewed
  • [Journal Article] A Survey of Sim-to-Real Transfer Techniques Applied to Reinforcement Learning for Bioinspired Robots2021

    • Author(s)
      Zhu Wei、Guo Xian、Owaki Dai、Kutsuzawa Kyo、Hayashibe Mitsuhiro
    • Journal Title

      IEEE Transactions on Neural Networks and Learning Systems

      Volume: 2021 Issue: 7 Pages: 1-16

    • DOI

      10.1109/tnnls.2021.3112718

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Int'l Joint Research
  • [Presentation] Learn to Navigate in Dynamic Environments with Normalized LiDAR Scans2024

    • Author(s)
      W. Zhu, M. Hayashibe
    • Organizer
      IEEE Int. Conf. on Robotics and Automation
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Learnable Tegotae-based Feedback in CPGs with Sparse Observation Produces Efficient and Adaptive Locomotion2024

    • Author(s)
      C. Herneth, M. Hayashibe, D. Owaki
    • Organizer
      IEEE Int. Conf. on Robotics and Automation
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Deep Reinforcement Learning for Tailorable Natural Quadruped Gait Generation2023

    • Author(s)
      L. Sulpice, D. Owaki, M. Hayashibe
    • Organizer
      11th Int. Symposium on Adaptive Motion of Animals and Mechanics
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] 機械学習を用いた物理振子群における同期現象の予測と評価2023

    • Author(s)
      田中裕人, 沓澤京, 大脇大, 林部充宏
    • Organizer
      計測自動制御学会 東北支部 第341回研究集会
    • Related Report
      2023 Annual Research Report
  • [Presentation] スパイク形式による画像の潜在表現を用いたモデルベース強化学習の性能評価2023

    • Author(s)
      平野貴也, 沓澤京, 大脇大, 林部充宏
    • Organizer
      ロボティクス・メカトロニクス講演会(ROBOMECH2023)
    • Related Report
      2023 Annual Research Report
  • [Presentation] 機械学習を用いた物理振子群の同期ダイナミクス予測2023

    • Author(s)
      田中裕人, 沓澤京, 大脇大, 林部充宏
    • Organizer
      ロボティクス・メカトロニクス講演会(ROBOMECH2023)
    • Related Report
      2023 Annual Research Report
  • [Presentation] Deep Reinforcement Learning based Robot Navigation in Dynamic Environments with Raw Laser Observations2023

    • Author(s)
      W. Zhu, M. Hayashibe
    • Organizer
      第41回日本ロボット学会学術講演会
    • Related Report
      2023 Annual Research Report
  • [Presentation] スパイク形式による画像の潜在表現を用いたモデルベース強化学習の性能評価2023

    • Author(s)
      平野貴也, 沓澤京, 大脇大, 林部充宏
    • Organizer
      計測自動制御学会 東北支部 第344回研究集会
    • Related Report
      2023 Annual Research Report
  • [Presentation] 二輪脚倒立振子型ロボットにおける膝関節制御の導入による不整地走破性および安定性の向上2023

    • Author(s)
      利根川太, 沓澤京, 大脇大, 林部充宏
    • Organizer
      第24回計測自動制御学会SI部門講演会
    • Related Report
      2023 Annual Research Report
  • [Presentation] Game-Based Evaluation of Whole-Body Movement Functions with CoM Stability and Motion Smoothness2022

    • Author(s)
      M. Kojima, K. Kutsuzawa, D. Owaki, M. Hayashibe
    • Organizer
      44th Annual Int. Conf. of the IEEE Engineering in Medicine and Biology Society
    • Related Report
      2022 Research-status Report
    • Int'l Joint Research
  • [Presentation] Systematic Motion Integration with Multiple Depth Cameras Allowing Sensor Movement for Stable Skeleton Tracking2022

    • Author(s)
      K. Furuhata, K. Kutsuzawa, D. Owaki, M. Hayashibe
    • Organizer
      44th Annual Int. Conf. of the IEEE Engineering in Medicine and Biology Society
    • Related Report
      2022 Research-status Report
    • Int'l Joint Research
  • [Presentation] Temporal Variation Quantification During Cognitive Dual-Task Gait Using Two IMU Sensors2022

    • Author(s)
      Y. Hutabarat, D. Owaki, M. Hayashibe
    • Organizer
      44th Annual Int. Conf. of the IEEE Engineering in Medicine and Biology Society
    • Related Report
      2022 Research-status Report
    • Int'l Joint Research
  • [Presentation] Classification of Human Balance Recovery Strategies through Kinematic Motor Synergy Analysis2022

    • Author(s)
      K. Shen, A. Chemori, M. Hayashibe
    • Organizer
      44th Annual Int. Conf. of the IEEE Engineering in Medicine and Biology Society
    • Related Report
      2022 Research-status Report
    • Int'l Joint Research
  • [Presentation] Reinforcement Learning based Hierarchical Control for Path Tracking of a Wheeled Bipedal Robot with Sim-to-Real Framework2022

    • Author(s)
      Zhu Wei、Raza Fahad、Hayashibe Mitsuhiro
    • Organizer
      IEEE/SICE International Symposium on System Integration (SII)
    • Related Report
      2021 Research-status Report
  • [Remarks] Neuro-Robotics Lab

    • URL

      http://neuro.mech.tohoku.ac.jp/

    • Related Report
      2023 Annual Research Report 2021 Research-status Report
  • [Remarks] Neuro-Robotics Lab, Tohoku University

    • URL

      http://neuro.mech.tohoku.ac.jp/

    • Related Report
      2022 Research-status Report

URL: 

Published: 2021-03-19   Modified: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi