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

Synthetic Robotics Approaches for Understanding the Formation and Deformation of Cooperation between Agents Based on Predictive Coding

Research Project

Project/Area Number 19K20364
Research Category

Grant-in-Aid for Early-Career Scientists

Allocation TypeMulti-year Fund
Review Section Basic Section 61040:Soft computing-related
Research InstitutionKeio University (2020-2022)
National Institute of Informatics (2019)

Principal Investigator

Murata Shingo  慶應義塾大学, 理工学部(矢上), 講師 (80778168)

Project Period (FY) 2019-04-01 – 2023-03-31
Project Status Completed (Fiscal Year 2022)
Budget Amount *help
¥4,030,000 (Direct Cost: ¥3,100,000、Indirect Cost: ¥930,000)
Fiscal Year 2021: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2020: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2019: ¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
Keywords予測符号化 / 予測誤差最小化 / 認知ロボティクス / ニューラルネットワーク / インタラクション
Outline of Research at the Start

本研究は他者との協調を支える認知情報処理機構の理解を目的とし,構成論的手法により取り組む.特に,(i)環境変化や他者のふるまいといった外的要因と(ii)自己の将来の行動に関する計画や意図といった内的要因により生じる協調の形成とその崩壊に関する動的過程に着目する.具体的には,脳の情報処理の仕組みとして提案されている予測符号化を再帰型神経回路モデルにより具現化し,二台のロボットそれぞれに実装する.そして,実環境における二台のロボット間の相互作用学習実験を行う.他者(ロボット)との相互作用の結果生じる互いの意図の動的な収斂と発散により,協調の形成・崩壊に関する動的過程が観察可能であると期待される.

Outline of Final Research Achievements

The aim of this research project is to understand the cognitive information processing mechanisms that support cooperation with others through synthetic robotics approaches integrating cognitive neuroscience, machine learning, and robotics. In particular, the focus is on the dynamic processes of the formation and deformation of cooperation, which are influenced by external factors such as other agents and the environment, as well as internal factors such as one's own intentions and goals. We have developed computational models based on predictive coding and conducted a set of robot experiments. Specifically, we have proposed a gradient-based optimization method and a more accelerated amortized inference method. We validated deep generative models equipped with these methods and applied them to collaborative robots to evaluate their performance in terms of both fundamental and practical aspects.

Academic Significance and Societal Importance of the Research Achievements

本研究成果は,他者や環境といった外的要因と自己の意図やゴールといった内的要因によって生じる協調の形成とその崩壊が,予測符号化という単一の仕組みによって説明可能であることを示した.具体的には,過去に生じた予測誤差を最小化することで外的要因を知覚し,未来に生じると想定される予測誤差を最小化することで内的要因を満たす行動生成が可能であることが確認された.本研究で得られた成果は他者との協調のみならず,共同注意や心の理論といった社会性認知に関する問題への貢献も期待される.また,ロボットを含む機械による他者(人間や他の機械)との円滑な協調の実現へと繋がる工学的応用可能性も期待される.

Report

(5 results)
  • 2022 Annual Research Report   Final Research Report ( PDF )
  • 2021 Research-status Report
  • 2020 Research-status Report
  • 2019 Research-status Report
  • Research Products

    (10 results)

All 2023 2022 2021 2020 2019

All Journal Article (4 results) (of which Peer Reviewed: 3 results,  Open Access: 1 results) Presentation (6 results) (of which Int'l Joint Research: 1 results)

  • [Journal Article] Deep Predictive Network for Inference and Dynamic Optimization of Task Goals during Human-Robot Collaboration2023

    • Author(s)
      Hiramatsu Shun, Murata Shingo
    • Journal Title

      Proceedings of the International Joint Conference on Neural Networks (IJCNN) 2023

      Volume: -

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Application of Robotic Predicitve Learning to Computational Psychiatry2022

    • Author(s)
      Murata Shingo
    • Journal Title

      Journal of the Robotics Society of Japan

      Volume: 40 Issue: 9 Pages: 796-801

    • DOI

      10.7210/jrsj.40.796

    • ISSN
      0289-1824, 1884-7145
    • Related Report
      2022 Annual Research Report
  • [Journal Article] Paradoxical sensory reactivity induced by functional disconnection in a robot model of neurodevelopmental disorder2021

    • Author(s)
      Idei Hayato、Murata Shingo、Yamashita Yuichi、Ogata Tetsuya
    • Journal Title

      Neural Networks

      Volume: 138 Pages: 150-163

    • DOI

      10.1016/j.neunet.2021.01.033

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access
  • [Journal Article] Looking Back and Ahead: Adaptation and Planning by Gradient Descent2019

    • Author(s)
      Murata Shingo,Sawa Hiroki,Sugano Shigeki,Ogata Tetsuya
    • Journal Title

      Proceedings of the Ninth Joint IEEE International Conference on Development and Learning and on Epigenetic Robotics (ICDL-EpiRob 2019)

      Volume: - Pages: 151-156

    • DOI

      10.1109/devlrn.2019.8850693

    • Related Report
      2019 Research-status Report
    • Peer Reviewed
  • [Presentation] Action Modification Based on Real-time Amortized Inference of Others' Intentions Using Backward RNN2022

    • Author(s)
      Orui Yukiko, Murata Shingo
    • Organizer
      The 54th ISCIE International Symposium on Stochastic Systems Theory and Its Applications (SSS '22)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Backward RNNを用いた階層的な潜在状態の動的推論2022

    • Author(s)
      大類有紀子, 村田真悟
    • Organizer
      日本発達神経科学会 第11回学術集会
    • Related Report
      2022 Annual Research Report
  • [Presentation] Backward RNNを用いた他者意図のリアルタイム推論に基づく行動修正の実現2022

    • Author(s)
      大類有紀子, 村田真悟
    • Organizer
      第40回日本ロボット学会学術講演会
    • Related Report
      2022 Annual Research Report
  • [Presentation] 深層学習を用いた協働ロボットによる作業目標の推論と動的修正の実現2022

    • Author(s)
      平松駿, 村田真悟
    • Organizer
      人工知能学会全国大会2022
    • Related Report
      2022 Annual Research Report
  • [Presentation] 過去から未来までの文脈を考慮した神経回路モデルによるロボットの目標に基づいた柔軟な行動生成2020

    • Author(s)
      佐藤琢,村田真悟,出井勇人,尾形哲也
    • Organizer
      人工知能学会全国大会2020
    • Related Report
      2020 Research-status Report
  • [Presentation] RNNを用いた予測不確実性と予測変化に基づく好奇心による行動選択モデルの提案2020

    • Author(s)
      樋園翼,斎藤菜美子,森裕紀,村田真悟,出井勇人,尾形哲也,菅野重樹
    • Organizer
      発達神経科学会 第9回学術集会
    • Related Report
      2020 Research-status Report

URL: 

Published: 2019-04-18   Modified: 2024-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi