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

Adaptive Probabilistic Robotics through Statistical Motion Analysis and Kinematics

Research Project

Project/Area Number 18H03295
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeSingle-year Grants
Section一般
Review Section Basic Section 61030:Intelligent informatics-related
Research InstitutionThe Institute of Statistical Mathematics

Principal Investigator

Mochihashi Daichi  統計数理研究所, 数理・推論研究系, 准教授 (80418508)

Co-Investigator(Kenkyū-buntansha) 高野 渉  大阪大学, 数理・データ科学教育研究センター, 特任教授(常勤) (30512090)
中村 友昭  電気通信大学, 大学院情報理工学研究科, 准教授 (50723623)
小林 一郎  お茶の水女子大学, 基幹研究院, 教授 (60281440)
Project Period (FY) 2018-04-01 – 2021-03-31
Project Status Completed (Fiscal Year 2020)
Budget Amount *help
¥16,510,000 (Direct Cost: ¥12,700,000、Indirect Cost: ¥3,810,000)
Fiscal Year 2020: ¥3,120,000 (Direct Cost: ¥2,400,000、Indirect Cost: ¥720,000)
Fiscal Year 2019: ¥6,370,000 (Direct Cost: ¥4,900,000、Indirect Cost: ¥1,470,000)
Fiscal Year 2018: ¥7,020,000 (Direct Cost: ¥5,400,000、Indirect Cost: ¥1,620,000)
Keywordsロボティクス / 自然言語処理 / 分節化 / ガウス過程 / 微分方程式 / 変化点検出 / トピックモデル / 隠れセミマルコフモデル / 動力学 / ベイズ統計 / ノンパラメトリックベイズ法 / 隠れマルコフモデル / 深層学習 / 統計モデル
Outline of Final Research Achievements

We conducted research with (a) statistical modeling of motions, and (b) statistical connection to natural language processing. As to (a), we extended motion segmentation with Gaussian processes for high-dimensional observations by conducting segmentation in a latent space induced by VAE, named HVGH. This research appeared at an international top conference of robotics (IROS) and an international journal.
As to (b), we developed a statistical model to explain adverbs that have been difficult to deal with in ordinary natural language processing. By learning a topic model on the space of functions induced by kernels, we succeeded to connect adverbs with associated motions toward adverb-driven motion generation.
During this research, the principal investigator authored a textbook "Gaussian processes and Machine Learning" from Kodansha, which sold over 10K copies to make Gaussian processes reachable to general audiences in machine learning.

Academic Significance and Societal Importance of the Research Achievements

開発した技術は、ロボットや人間の動作の「形態素解析」にあたる基礎的な研究であり、これにより連続的な動作時系列を単語を数えるように統計的に分析することが可能になった。特に、観測値が各関節角で高次元な場合でも扱えるHVGHを開発したことで、潜在空間での分節化が行えるようになったことは、統計モデルとしても重要な進歩であるといえる。
また、副詞はロボットとのインタラクションで重要な働きをすると予想されるが、従来の自然言語処理では適切に扱うことが難しかった。本研究により関数空間のトピックモデルとして統計モデル化できたことで、「慎重に運んで」「さっと拭いて」などの、より適切な動作が可能なロボットに繋がる。

Report

(4 results)
  • 2020 Annual Research Report   Final Research Report ( PDF )
  • 2019 Annual Research Report
  • 2018 Annual Research Report
  • Research Products

    (23 results)

All 2021 2020 2019 2018

All Journal Article (6 results) (of which Peer Reviewed: 5 results,  Open Access: 4 results) Presentation (15 results) (of which Int'l Joint Research: 5 results,  Invited: 5 results) Book (2 results)

  • [Journal Article] MC-GP-HSMMを用いたマルチモーダル情報の分節化によるインタラクションのルール学習2021

    • Author(s)
      木村大河,長野匡隼,中村友昭
    • Journal Title

      日本ロボット学会誌

      Volume: -

    • NAID

      130008069768

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Motion Generation Using Humanoid Robot with Language Understanding2020

    • Author(s)
      濱園侑美, 小林一郎, 麻生英樹, 中村友昭, 長井隆行, 持橋大地
    • Journal Title

      Journal of Japan Society for Fuzzy Theory and Intelligent Informatics

      Volume: 32 Issue: 1 Pages: 632-642

    • DOI

      10.3156/jsoft.32.1_632

    • NAID

      130007798537

    • ISSN
      1347-7986, 1881-7203
    • Year and Date
      2020-02-15
    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] HVGH: Unsupervised Segmentation for High-dimensional Time Series Using Deep Neural Compression and Statistical Generative Mode2019

    • Author(s)
      Masatoshi Nagano, Tomoaki Nakamura, Takayuki Nagai, Daichi Mochihashi, Ichiro Kobayashi, Wataru Takano
    • Journal Title

      Frontiers in Robotics and AI

      Volume: 6 Pages: 1-15

    • DOI

      10.3389/frobt.2019.00115

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Survey on frontiers of language and robotics2019

    • Author(s)
      Tangiuchi T.、Mochihashi D.、Nagai T.、Uchida S.、Inoue N.、Kobayashi I.、Nakamura T.、Hagiwara Y.、Iwahashi N.、Inamura T.
    • Journal Title

      Advanced Robotics

      Volume: 33 Issue: 15-16 Pages: 700-730

    • DOI

      10.1080/01691864.2019.1632223

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Sequential Monte Carlo Controller that Integrates Physical Consistency and Motion Knowledge2019

    • Author(s)
      Wataru Takano, Taro Takahashi, Yoshihiko Nakamura
    • Journal Title

      Autonomous Robots

      Volume: -

    • Related Report
      2018 Annual Research Report
  • [Journal Article] Sequence Pattern Extraction by Segmenting Time Series Data Using GP-HSMM with Hierarchical Dirichlet Process2018

    • Author(s)
      Masatoshi Nagano, Tomoaki Nakamura, Takayuki Nagai, Daichi Mochihashi, Ichiro Kobayashi, Masahide Kaneko
    • Journal Title

      IROS 2018

      Volume: 0 Pages: 4067-4074

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed / Open Access
  • [Presentation] スペクトル混合カーネルとガウス過程に基づく動画からの副詞の意味理解2021

    • Author(s)
      谷口巴, 持橋大地, 長野匡隼, 中村友昭, 長井隆行, 高野渉, 小林一郎
    • Organizer
      言語処理学会2021 P6-17
    • Related Report
      2020 Annual Research Report
  • [Presentation] 畳み込み変分オートエンコーダとガウス過程に基づく動画像の分節化2021

    • Author(s)
      長野匡隼,中村友昭,長井隆行,持橋大地,小林一郎,高野渉
    • Organizer
      人工知能学会全国大会 2J3-GS-8b-01
    • Related Report
      2020 Annual Research Report
  • [Presentation] ガウス過程と自然言語処理2021

    • Author(s)
      持橋大地
    • Organizer
      言語処理学会2021 チュートリアルT1
    • Related Report
      2020 Annual Research Report
    • Invited
  • [Presentation] MC-GP-HSMMを用いたマルチモーダル情報の分節化によるインタラクションのルール学習2020

    • Author(s)
      木村大河,長野匡隼,中村友昭
    • Organizer
      日本ロボット学会学術講演会,1C2-04
    • Related Report
      2020 Annual Research Report
  • [Presentation] Spectral Mixture Kernelを用いた動作を表す副詞の意味理解へ向けた取り組み2020

    • Author(s)
      谷口巴,長野匡隼,持橋大地,中村友昭,高野渉,長井隆行,小林一郎
    • Organizer
      人工知能学会全国大会,1Q5-GS-11-02
    • Related Report
      2020 Annual Research Report
  • [Presentation] High-dimensional Motion Segmentation by Variational Autoencoder and Gaussian Processes2019

    • Author(s)
      Masatoshi Nagano, Tomoaki Nakamura, Takayuki Nagai, Daichi Mochihashi, Ichiro Kobayashi, Wataru Takano
    • Organizer
      IROS 2019
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research
  • [Presentation] HVGH: 高次元時系列データの深層圧縮と教師なし分節化2019

    • Author(s)
      長野匡隼, 中村友昭, 長井隆行, 持橋大地, 小林一郎, 高野渉
    • Organizer
      2019年度人工知能学会全国大会 1L3-J-11-01
    • Related Report
      2019 Annual Research Report
  • [Presentation] 霊長類における身体動作時系列の分節推移構造推定2019

    • Author(s)
      三村喬生, 中村友昭 松本惇平, 西条寿夫, 須原哲也, 持橋大地, 南本敬史
    • Organizer
      2019年度人工知能学会全国大会 1C4-J-3-01
    • Related Report
      2019 Annual Research Report
  • [Presentation] High-dimensional motion segmentation with semi-Markov Latent Gaussian Processes2019

    • Author(s)
      Daichi Mochihashi
    • Organizer
      ISBA East Asian Chapter
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] Gaussian processes for recognizing Motions in robots2019

    • Author(s)
      Daichi Mochihashi
    • Organizer
      計測自動制御学会 SICE 2019
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] Gaussian Process Generative Models for Language and Robotics2019

    • Author(s)
      Daichi Mochihashi
    • Organizer
      CoRL 2019 Tutorial
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] High-dimensional motion segmentation with semi-Markov Latent Gaussian processes2019

    • Author(s)
      Daichi Mochihashi
    • Organizer
      University of Bristol, Jean Golding Institute, "High Dimensional and Bayesian Inference toward Quantifying Real-World Uncertainties" Workshop
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] VAEとガウス過程による高次元データの圧縮と同時分節化2018

    • Author(s)
      長野匡隼,中村友昭,長井隆行,持橋大地,小林一郎,高野渉,金子正秀
    • Organizer
      情報論的学習理論ワークショップ,D1-24
    • Related Report
      2018 Annual Research Report
  • [Presentation] ノンパラメトリックベイズ法に基づく時系列データの分節化2018

    • Author(s)
      長野匡隼,中村友昭,長井隆行,持橋大地,小林一郎,金子 正秀
    • Organizer
      人工知能学会全国大会,2G4-04
    • Related Report
      2018 Annual Research Report
  • [Presentation] 階層ディリクレ過程による動作クラス数推定を導入したGP-HSMMによる連続動作からの基本動作抽出2018

    • Author(s)
      長野匡隼,中村友昭,長井隆行,持橋大地,小林一郎,金子正秀
    • Organizer
      情報処理学会全国大会,6M-03
    • Related Report
      2018 Annual Research Report
  • [Book] ガウス過程と機械学習2019

    • Author(s)
      持橋大地, 大羽成征
    • Total Pages
      256
    • Publisher
      講談社サイエンティフィク
    • ISBN
      4061529269
    • Related Report
      2019 Annual Research Report
  • [Book] ガウス過程と機械学習2019

    • Author(s)
      持橋大地, 大羽成征
    • Total Pages
      256
    • Publisher
      講談社
    • ISBN
      9784061529267
    • Related Report
      2018 Annual Research Report

URL: 

Published: 2018-04-23   Modified: 2022-01-27  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi