OGATA, Tetsuya

写真a

Affiliation

Faculty of Science and Engineering, School of Fundamental Science and Engineering

Job title

Professor

Homepage URL

http://ogata-lab.jp

Profile

Tetsuya Ogata received the B.S., M.S., and D.E. degrees in mechanical engineering from Waseda University, Tokyo, Japan, in 1993, 1995, and 2000, respectively. He was a Research Associate with Waseda University from 1999 to 2001. From 2001 to 2003, he was a Research Scientist with the RIKEN Brain Science Institute, Saitama, Japan. From 2003 to 2012, he was an Associate Professor with the Graduate School of Informatics, Kyoto University, Kyoto, Japan. Since 2012, he has been a Professor with the Faculty of Science and Engineering, Waseda University. From 2009 to 2015, he was a JST (Japan Science and Technology Agency) PREST Researcher. From 2017, he is a Joint-appointed Fellow with the Artificial Intelligence Research Center, National Institute of Advanced Industrial Science and Technology, Tokyo. He served as director of the Robotics Society of Japan (RSJ) from 2014 to 2015 and of the Japanese Society of Artificial Intelligence (JSAI) from 2016 to 2018. He is currently a director of Japan Deep Learning Association (JDLA) from 2017. His current research interests include deep learning for robot motion control, human–robot interaction, and dynamics of human–robot mutual adaptation.

Concurrent Post 【 display / non-display

  • Affiliated organization   Global Education Center

  • Faculty of Science and Engineering   Graduate School of Fundamental Science and Engineering

Research Institute 【 display / non-display

  • 2020
    -
    2022

    理工学術院総合研究所   兼任研究員

Education 【 display / non-display

  • 1995.04
    -
    2000.03

    Waseda University   Graduate School of Science and Engineering   Department of Mechanical Engineering  

    博士後期課程

  • 1993.04
    -
    1995.03

    Waseda University   Graduate School of Science and Engineering   Department of Mechanical Engineering  

    Master Course

  • 1989.04
    -
    1993.03

    Waseda University   School of Science and Engineering   Department of Mechanical Engineering  

Degree 【 display / non-display

  • 早稲田大学   博士(工学)

  • Waseda University   Ph.D, Engineering

Research Experience 【 display / non-display

  • 2017.10
    -
     

    National Institute of Advanced Industrial Science and Technology (AIST)   Artificial Intelligence Research Center (AIRC)   Joint-appointed Fellow

  • 2009.10
    -
    2015.03

    Japan Science and Technology Agency   PRESTO Researcher

  • 2005.06
    -
    2012.03

    Kyoto University   情報学研究科   Associate Professor

  • 2005.06
    -
    2012.03

    Kyoto University   情報学研究科   Associate Professor

  • 2003.10
    -
    2005.05

    Kyoto University   Graduate School of Informatics   Lecturer

display all >>

Professional Memberships 【 display / non-display

  •  
     
     

    IEEE

  •  
     
     

    The Japanese Society for Artificial Intelligence

  •  
     
     

    The Society of Instrument and Control Engieers

  •  
     
     

    Society of Biomechanism Japan

  •  
     
     

    The Japan Society of Mechanical Engineering

display all >>

 

Research Areas 【 display / non-display

  • Intelligent robotics

Research Interests 【 display / non-display

  • Cognitive Robotics

Papers 【 display / non-display

  • Development of a basic educational kit for robotic system with deep neural networks

    Momomi Kanamura, Kanata Suzuki, Yuki Suga, Tetsuya Ogata

    Sensors   21 ( 11 )  2021.06

     View Summary

    In many robotics studies, deep neural networks (DNNs) are being actively studied due to their good performance. However, existing robotic techniques and DNNs have not been systematically integrated, and packages for beginners are yet to be developed. In this study, we proposed a basic educational kit for robotic system development with DNNs. Our goal was to educate beginners in both robotics and machine learning, especially the use of DNNs. Initially, we required the kit to (1) be easy to understand, (2) employ experience-based learning, and (3) be applicable in many areas. To clarify the learning objectives and important parts of the basic educational kit, we analyzed the research and development (R&D) of DNNs and divided the process into three steps of data collection (DC), machine learning (ML), and task execution (TE). These steps were configured under a hierarchical system flow with the ability to be executed individually at the development stage. To evaluate the practicality of the proposed system flow, we implemented it for a physical robotic grasping system using robotics middleware. We also demonstrated that the proposed system can be effectively applied to other hardware, sensor inputs, and robot tasks.

    DOI PubMed

  • Paradoxical sensory reactivity induced by functional disconnection in a robot model of neurodevelopmental disorder

    Hayato Idei, Shingo Murata, Yuichi Yamashita, Tetsuya Ogata

    Neural Networks   138   150 - 163  2021.06  [Refereed]

     View Summary

    Neurodevelopmental disorders are characterized by heterogeneous and non-specific nature of their clinical symptoms. In particular, hyper- and hypo-reactivity to sensory stimuli are diagnostic features of autism spectrum disorder and are reported across many neurodevelopmental disorders. However, computational mechanisms underlying the unusual paradoxical behaviors remain unclear. In this study, using a robot controlled by a hierarchical recurrent neural network model with predictive processing and learning mechanism, we simulated how functional disconnection altered the learning process and subsequent behavioral reactivity to environmental change. The results show that, through the learning process, long-range functional disconnection between distinct network levels could simultaneously lower the precision of sensory information and higher-level prediction. The alteration caused a robot to exhibit sensory-dominated and sensory-ignoring behaviors ascribed to sensory hyper- and hypo-reactivity, respectively. As long-range functional disconnection became more severe, a frequency shift from hyporeactivity to hyperreactivity was observed, paralleling an early sign of autism spectrum disorder. Furthermore, local functional disconnection at the level of sensory processing similarly induced hyporeactivity due to low sensory precision. These findings suggest a computational explanation for paradoxical sensory behaviors in neurodevelopmental disorders, such as coexisting hyper- and hypo-reactivity to sensory stimulus. A neurorobotics approach may be useful for bridging various levels of understanding in neurodevelopmental disorders and providing insights into mechanisms underlying complex clinical symptoms.

    DOI PubMed

  • From Anime to Reality: Embodying An Anime Character As A Humanoid Robot

    Mohammed Al Al Sada, Pin Chu Yang, Chang Chieh Chiu, Tito Pradhono Tomo, Mhd Yamen Saraiji, Tetsuya Ogata, Tatsuo Nakajima

    Conference on Human Factors in Computing Systems - Proceedings    2021.05

     View Summary

    Otaku is a Japanese term commonly associated with fans of Japanese animation, comics or video games. Otaku culture has grown to be a global phenomenon with various hobbies and media. Despite its popularity, research efforts to contribute to the otaku culture have been modest. Therefore, we present Hatsuki, which is a humanoid robot that is especially designed to embody anime characters. Hatsuki advances the state of the art as it: 1) realizes aesthetics resembling anime characters, 2) implements 2D anime-like facial expression system, and 3) realizes anime-style behaviors and interactions. We explain Hatsuki's design specifics and its interaction domains as an autonomous robot and as a teleoperated humanoid avatar. We discuss our efforts under each interaction domain, and follow by discussing its potential deployment venues and applications. We highlight opportunities of interplay between otaku culture and interactive systems, potentially enabling highly desirable interactions and familiar system designs to users exposed to otaku culture.

    DOI

  • Embodying Pre-Trained Word Embeddings through Robot Actions

    Minori Toyoda, Kanata Suzuki, Hiroki Mori, Yoshihiko Hayashi, Tetsuya Ogata

    IEEE Robotics and Automation Letters   6 ( 2 ) 4225 - 4232  2021.04

     View Summary

    We propose a promising neural network model with which to acquire a grounded representation of robot actions and the linguistic descriptions thereof. Properly responding to various linguistic expressions, including polysemous words, is an important ability for robots that interact with people via linguistic dialogue. Previous studies have shown that robots can use words that are not included in the action-description paired datasets by using pre-trained word embeddings. However, the word embeddings trained under the distributional hypothesis are not grounded, as they are derived purely from a text corpus. In this letter, we transform the pre-trained word embeddings to embodied ones by using the robot's sensory-motor experiences. We extend a bidirectional translation model for actions and descriptions by incorporating non-linear layers that retrofit the word embeddings. By training the retrofit layer and the bidirectional translation model alternately, our proposed model is able to transform the pre-trained word embeddings to adapt to a paired action-description dataset. Our results demonstrate that the embeddings of synonyms form a semantic cluster by reflecting the experiences (actions and environments) of a robot. These embeddings allow the robot to properly generate actions from unseen words that are not paired with actions in a dataset.

    DOI

  • Compensation for Undefined Behaviors during Robot Task Execution by Switching Controllers Depending on Embedded Dynamics in RNN

    Kanata Suzuki, Hiroki Mori, Tetsuya Ogata

    IEEE Robotics and Automation Letters   6 ( 2 ) 3475 - 3482  2021.04  [Refereed]

    Authorship:Last author, Corresponding author

     View Summary

    Robotic applications require both correct task performance and compensation for undefined behaviors. Although deep learning is a promising approach to perform complex tasks, the response to undefined behaviors that are not reflected in the training dataset remains challenging. In a human-robot collaborative task, the robot may adopt an unexpected posture due to collisions and other unexpected events. Therefore, robots should be able to recover from disturbances for completing the execution of the intended task. We propose a compensation method for undefined behaviors by switching between two controllers. Specifically, the proposed method switches between learning-based and model-based controllers depending on the internal representation of a recurrent neural network that learns task dynamics. We applied the proposed method to a pick-And-place task and evaluated the compensation for undefined behaviors. Experimental results from simulations and on a real robot demonstrate the effectiveness and high performance of the proposed method.

    DOI

display all >>

Books and Other Publications 【 display / non-display

  • 「こころ」とアーティフィシャル・マインド

    河合, 俊雄, 吉岡, 洋, 西垣, 通, 尾形, 哲也, 長尾, 真

    創元社  2021.02 ISBN: 9784422117577

  • 発達ロボティクスハンドブック : ロボットで探る認知発達の仕組み

    Cangelosi, Angelo, Schlesinger, Matthew, 萩原, 良信, 荒川, 直哉, 長井, 隆行, 尾形, 哲也, 稲邑, 哲也, 岩橋, 直人, 杉浦, 孔明, 牧野, 武文, 岡田, 浩之, 谷口, 忠大

    福村出版  2019.01 ISBN: 9784571230592

  • ディープラーニングがロボットを変える

    尾形, 哲也

    日刊工業新聞社  2017.07 ISBN: 9784526077326

Misc 【 display / non-display

  • Deep Neural Networkを用いたマルチモーダル音声認識

    野田邦昭, 山口雄紀, 中臺一博, 奥乃博, 尾形哲也

    日本ロボット学会学術講演会予稿集(CD-ROM)   32nd   ROMBUNNO.1I1-04  2014.09

    J-GLOBAL

  • 人間の描画発達に基づくロボットの描画模倣学習モデルの構築

    西出俊, 望月敬太, 奥乃博, 尾形哲也

    日本ロボット学会学術講演会予稿集(CD-ROM)   32nd   ROMBUNNO.2I2-04  2014.09

    J-GLOBAL

  • マイク数以上の同時発話分離のための調波・非調波音源モデルの検討

    平澤恭治, 安良岡直希, 高橋徹, 尾形哲也, 奥乃博

    第74回全国大会講演論文集   2012 ( 1 ) 577 - 578  2012.03

     View Summary

    人間の生活環境に存在する多数の音源を正しく分離するために, 我々はマイク数以上の音源分離(劣決定音源分離)の検討を行っている. 我々は同時発話分離のために混合ガウス分布を用いた調波・非調波音源モデルを提案しているが, 調波部分は理論的に正当化されたモデル化が行われている一方で, 非調波部分のモデルには理論的な正当性がなく, 分離性能を低下させる要因となっていた. 本稿では従来の調波・非調波音源モデルに対して変更を加え, SiSEC2011で用いられた男女3-4話者の混合音声を分離する実験により, どのようなモデルが劣決定同時発話分離問題に適しているかを検討する.

    CiNii

  • マイク数以上の同時発話分離のための調波・非調波音源モデルの検討

    平澤恭治, 安良岡直希, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会全国大会講演論文集   74th ( 2 ) 2.577-2.578 - 579  2012.03

     View Summary

    人間の生活環境に存在する多数の音源を正しく分離するために, 我々はマイク数以上の音源分離(劣決定音源分離)の検討を行っている. 我々は同時発話分離のために混合ガウス分布を用いた調波・非調波音源モデルを提案しているが, 調波部分は理論的に正当化されたモデル化が行われている一方で, 非調波部分のモデルには理論的な正当性がなく, 分離性能を低下させる要因となっていた. 本稿では従来の調波・非調波音源モデルに対して変更を加え, SiSEC2011で用いられた男女3-4話者の混合音声を分離する実験により, どのようなモデルが劣決定同時発話分離問題に適しているかを検討する.

    CiNii J-GLOBAL

  • 同時複数音源に対する擬音語による音源選択システム

    山村祐介, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会全国大会講演論文集   74th ( 2 ) 2.587-2.588  2012.03

    J-GLOBAL

display all >>

Industrial Property Rights 【 display / non-display

  • 情報処理システムおよび情報処理方法、並びにプログラム

    4472506

    林 隆志, 金 天海, 尾形 哲也

    Patent

Awards 【 display / non-display

  • Best paper award on Cognitive Robotics

    2021.06   IEEE International Conference on Robotics and Automation (ICRA 2021)   How to select and use tools? : Active Perception of Target Objects Using Multimodal Deep Learning

    Winner: Namiko Saito, Tetsuya Ogata, Satoshi Funabashi, Hiroki Mori, Shigeki Sugano

  • SI2020優秀講演賞

    2020.12   計測自動制御学会SI部門   データ変換に着目したミドルウェアモデルにおけるデバイス類の位置の扱いに関する議論

    Winner: 菅佑樹, 森裕紀, 尾形哲也

  • 全国大会優秀賞

    2020.07   人工知能学会   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

    Winner: 佐藤琢, 村田真悟, 出井勇人, 尾形哲也

  • 全国大会優秀賞

    2020.07   人工知能学会   未知語に対応可能な言語と動作の統合表現獲得モデル

    Winner: 豊田みのり, 森裕紀, 鈴木彼方, 林良彦, 尾形哲也

  • 論文賞

    2019.12   FA財団  

display all >>

Research Projects 【 display / non-display

  • 一人に一台一生寄り添うスマートロボットAIREC

    Project Year :

    2020.10
    -
    2025.03
     

    Authorship: Other

  • 実世界に埋め込まれる人間中心の人工知能技術の研究開発

    Project Year :

    2020.04
    -
    2022.03
     

    Authorship: Other

  • Dense 3-axis tactile sensing and AI to implement human-like manual skills in robots

    Grant-in-Aid for Scientific Research (B)

    Project Year :

    2019.04
    -
    2022.03
     

    シュミッツ アレクサンダー, 尾形 哲也, 玉城 絵美, Somlor Sophon

    Authorship: Coinvestigator(s)

     View Summary

    In this research we develop a smart sensing system that enables robot hands to achieve human-like manipulation skills. Key components are 1. dense 3-axis tactile sensors for robot hands and 2. learning algorithms exploiting massive 3-axis tactile data for intelligent force control.
    <BR>
    We integrated the tactile skin sensors in grippers and robot hands. Using a novel joint (with a remote center of motion mechanism) we could achieve full coverage of the palmar side of the fingers with sensors in one gripper. Furthermore, we instrumented human hands with the sensors, to enable skill transfer from human to robot hands in the future. We used the skin sensors integrated in the robot hands for various machine learning experiments. In particular, we used deep convolutional neural networks for tactile object recognition as well as for in-hand manipulation.

  • 記号創発ロボティクスによる人間機械コラボレーション基盤創成

    Project Year :

    2015.10
    -
    2021.03
     

    Authorship: Other

  • 日常生活支援ロボット

    Project Year :

    2017.04
    -
    2020.03
     

    Authorship: Other

display all >>

Presentations 【 display / non-display

display all >>

Specific Research 【 display / non-display

  • 階層型神経回路モデルにおける予測可能性を利用した自己身体モデルの獲得

    2015  

     View Summary

    本研究では,ロボットの学習機構の軸となる機構としての,自己身体モデルについて,特に視野内の自己領域と外部物体とを区別する基礎モデルを提案し,認知モデルとの対応と理解,及びロボットシステムへの応用を目指している.我々は特に,再起結合型神経回モデル(RNN)の一種であるStochastic ContinuousTime Recurrent Neural Network(S-CTRNN)を用いた方法を提案した. S-CTRNNは時系列変化の予測のみならず,その不確実性を分散として予測することが可能なモデルである.このS-CTRNNを人間型ロボットに実装し,視野内の自己のハンドとボールとのインタラクションを観察,学習させる実験を行った.その結果分散予測によって,自己身体と外部物体の運動の区別を行える可能性が示された.&nbsp;

 

Syllabus 【 display / non-display

display all >>

 

Committee Memberships 【 display / non-display

  • 2016.07
    -
    2018.06

    人工知能学会  理事

  • 2018.04
    -
     

    G1 Institute Deep Learning Research Group  Advisory Board Member

  • 2018.04
    -
     

    一般社団法人G1ディープラーニング研究会  アドバイザリー・ボードメンバー

  • 2018.04
    -
     

    The Society of Instrument and Control Engineers  Director

  • 2018.04
    -
     

    計測自動制御学会  理事

display all >>

Social Activities 【 display / non-display

  • Technical Advisor

    Integral AI, Inc. 

    2021
    -
    Now

  • ACT-X研究「AI活用学問革新」領域アドバイザー

    科学技術振興機構 

    2020
    -
    Now

  • アドバイザー

    株式会社アバターイン 

    2020
    -
    Now

  • アドバイザー

    IGPI テクノロジー 

    2018
    -
    Now

  • さきがけ研究「社会デザイン」領域アドバイザー

    科学技術振興機構 

    2017
    -
    Now

display all >>

Media Coverage 【 display / non-display

  • Robot learns to tie knots using only two fingers on each hand

    Internet

    Author: Other  

    NewScientist  

    2021.03

  • 神経回路モデル搭載ロボットで、ASDの認知行動異常を解明-早大ほか

    Internet

    Author: Other  

    医療NEWS  

    2020.08

  • 早大と国立精神・神経医療研究センター、神経発達障害の認知行動異常のメカニズムを解明

    Newspaper, magazine

    Author: Other  

    日本経済新聞  

    2020.08

  • ディープラーニングがロボットを多能にする,進化を続ける人工知能AI

    Newspaper, magazine

    日経サイエンス  

    p.96  

    2020.03

  • What is the Cutieroid project that develops life-size “moving figures”?

    Internet

    Gigazine  

    2020.02

display all >>

Academic Activities 【 display / non-display

  • Member of National Committee

    2004
    -
    Now
  • 欧文誌委員

    Academic society, research group, etc.

    日本ロボット学会  

    2004
    -
    Now
  • 企画委員

    Academic society, research group, etc.

    人工知能学会  

    2018
    -
    2020
  • 理事

    Academic society, research group, etc.

    計測自動制御学会  

    2018
    -
    2020
  • 会誌編集委員

    Academic society, research group, etc.

    人工知能学会  

    2015
    -
    2019

display all >>