Neural network based model for visual-motor integration learning of robot's drawing behavior: Association of a drawing motion from a drawn image

Kazuma Sasaki, Hadi Tjandra, Kuniaki Noda, Kuniyuki Takahashi, Tetsuya Ogata

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    9 Citations (Scopus)

    Abstract

    In this study, we propose a neural network based model for learning a robot's drawing sequences in an unsupervised manner. We focus on the ability to learn visual-motor relationships, which can work as a reusable memory in association of drawing motion from a picture image. Assuming that a humanoid robot can draw a shape on a pen tablet, the proposed model learns drawing sequences, which comprises drawing motion and drawn picture image frames. To learn raw pixel data without any given specific features, we utilized a deep neural network for compressing large dimensional picture images and a continuous time recurrent neural network for integration of motion and picture images. To confirm the ability of the proposed model, we performed an experiment for learning 15 sequences comprising three types of shapes. The model successfully learns all the sequences and can associate a drawing motion from a not trained picture image and a trained picture with similar success. We also show that the proposed model self-organizes its behavior according to types shapes.

    Original languageEnglish
    Title of host publicationIEEE International Conference on Intelligent Robots and Systems
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    Pages2736-2741
    Number of pages6
    Volume2015-December
    ISBN (Print)9781479999941
    DOIs
    Publication statusPublished - 2015 Dec 11
    EventIEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2015 - Hamburg, Germany
    Duration: 2015 Sep 282015 Oct 2

    Other

    OtherIEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2015
    CountryGermany
    CityHamburg
    Period15/9/2815/10/2

    Fingerprint

    Robots
    Neural networks
    Recurrent neural networks
    Pixels
    Data storage equipment
    Experiments

    Keywords

    • Context
    • Neurons
    • Recurrent neural networks
    • Robots
    • Shape
    • Training

    ASJC Scopus subject areas

    • Control and Systems Engineering
    • Software
    • Computer Vision and Pattern Recognition
    • Computer Science Applications

    Cite this

    Sasaki, K., Tjandra, H., Noda, K., Takahashi, K., & Ogata, T. (2015). Neural network based model for visual-motor integration learning of robot's drawing behavior: Association of a drawing motion from a drawn image. In IEEE International Conference on Intelligent Robots and Systems (Vol. 2015-December, pp. 2736-2741). [7353752] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IROS.2015.7353752

    Neural network based model for visual-motor integration learning of robot's drawing behavior : Association of a drawing motion from a drawn image. / Sasaki, Kazuma; Tjandra, Hadi; Noda, Kuniaki; Takahashi, Kuniyuki; Ogata, Tetsuya.

    IEEE International Conference on Intelligent Robots and Systems. Vol. 2015-December Institute of Electrical and Electronics Engineers Inc., 2015. p. 2736-2741 7353752.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Sasaki, K, Tjandra, H, Noda, K, Takahashi, K & Ogata, T 2015, Neural network based model for visual-motor integration learning of robot's drawing behavior: Association of a drawing motion from a drawn image. in IEEE International Conference on Intelligent Robots and Systems. vol. 2015-December, 7353752, Institute of Electrical and Electronics Engineers Inc., pp. 2736-2741, IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2015, Hamburg, Germany, 15/9/28. https://doi.org/10.1109/IROS.2015.7353752
    Sasaki K, Tjandra H, Noda K, Takahashi K, Ogata T. Neural network based model for visual-motor integration learning of robot's drawing behavior: Association of a drawing motion from a drawn image. In IEEE International Conference on Intelligent Robots and Systems. Vol. 2015-December. Institute of Electrical and Electronics Engineers Inc. 2015. p. 2736-2741. 7353752 https://doi.org/10.1109/IROS.2015.7353752
    Sasaki, Kazuma ; Tjandra, Hadi ; Noda, Kuniaki ; Takahashi, Kuniyuki ; Ogata, Tetsuya. / Neural network based model for visual-motor integration learning of robot's drawing behavior : Association of a drawing motion from a drawn image. IEEE International Conference on Intelligent Robots and Systems. Vol. 2015-December Institute of Electrical and Electronics Engineers Inc., 2015. pp. 2736-2741
    @inproceedings{de7b1f1428a84702a644f3c8099fa966,
    title = "Neural network based model for visual-motor integration learning of robot's drawing behavior: Association of a drawing motion from a drawn image",
    abstract = "In this study, we propose a neural network based model for learning a robot's drawing sequences in an unsupervised manner. We focus on the ability to learn visual-motor relationships, which can work as a reusable memory in association of drawing motion from a picture image. Assuming that a humanoid robot can draw a shape on a pen tablet, the proposed model learns drawing sequences, which comprises drawing motion and drawn picture image frames. To learn raw pixel data without any given specific features, we utilized a deep neural network for compressing large dimensional picture images and a continuous time recurrent neural network for integration of motion and picture images. To confirm the ability of the proposed model, we performed an experiment for learning 15 sequences comprising three types of shapes. The model successfully learns all the sequences and can associate a drawing motion from a not trained picture image and a trained picture with similar success. We also show that the proposed model self-organizes its behavior according to types shapes.",
    keywords = "Context, Neurons, Recurrent neural networks, Robots, Shape, Training",
    author = "Kazuma Sasaki and Hadi Tjandra and Kuniaki Noda and Kuniyuki Takahashi and Tetsuya Ogata",
    year = "2015",
    month = "12",
    day = "11",
    doi = "10.1109/IROS.2015.7353752",
    language = "English",
    isbn = "9781479999941",
    volume = "2015-December",
    pages = "2736--2741",
    booktitle = "IEEE International Conference on Intelligent Robots and Systems",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",

    }

    TY - GEN

    T1 - Neural network based model for visual-motor integration learning of robot's drawing behavior

    T2 - Association of a drawing motion from a drawn image

    AU - Sasaki, Kazuma

    AU - Tjandra, Hadi

    AU - Noda, Kuniaki

    AU - Takahashi, Kuniyuki

    AU - Ogata, Tetsuya

    PY - 2015/12/11

    Y1 - 2015/12/11

    N2 - In this study, we propose a neural network based model for learning a robot's drawing sequences in an unsupervised manner. We focus on the ability to learn visual-motor relationships, which can work as a reusable memory in association of drawing motion from a picture image. Assuming that a humanoid robot can draw a shape on a pen tablet, the proposed model learns drawing sequences, which comprises drawing motion and drawn picture image frames. To learn raw pixel data without any given specific features, we utilized a deep neural network for compressing large dimensional picture images and a continuous time recurrent neural network for integration of motion and picture images. To confirm the ability of the proposed model, we performed an experiment for learning 15 sequences comprising three types of shapes. The model successfully learns all the sequences and can associate a drawing motion from a not trained picture image and a trained picture with similar success. We also show that the proposed model self-organizes its behavior according to types shapes.

    AB - In this study, we propose a neural network based model for learning a robot's drawing sequences in an unsupervised manner. We focus on the ability to learn visual-motor relationships, which can work as a reusable memory in association of drawing motion from a picture image. Assuming that a humanoid robot can draw a shape on a pen tablet, the proposed model learns drawing sequences, which comprises drawing motion and drawn picture image frames. To learn raw pixel data without any given specific features, we utilized a deep neural network for compressing large dimensional picture images and a continuous time recurrent neural network for integration of motion and picture images. To confirm the ability of the proposed model, we performed an experiment for learning 15 sequences comprising three types of shapes. The model successfully learns all the sequences and can associate a drawing motion from a not trained picture image and a trained picture with similar success. We also show that the proposed model self-organizes its behavior according to types shapes.

    KW - Context

    KW - Neurons

    KW - Recurrent neural networks

    KW - Robots

    KW - Shape

    KW - Training

    UR - http://www.scopus.com/inward/record.url?scp=84958173984&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84958173984&partnerID=8YFLogxK

    U2 - 10.1109/IROS.2015.7353752

    DO - 10.1109/IROS.2015.7353752

    M3 - Conference contribution

    AN - SCOPUS:84958173984

    SN - 9781479999941

    VL - 2015-December

    SP - 2736

    EP - 2741

    BT - IEEE International Conference on Intelligent Robots and Systems

    PB - Institute of Electrical and Electronics Engineers Inc.

    ER -