Applying intrinsic motivation for visuomotor learning of robot arm motion

Shun Nishide, Harumitsu Nobuta, Hiroshi G. Okuno, Tetsuya Ogata

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

    Abstract

    In this paper, we present a method to apply intrinsic motivation for improving visuomotor learning of robot's arm with external object in view. Multiple Timescales Recurrent Neural Network (MTRNN) is utilized for learning the robot arm/external object dynamics. Training of MTRNN is done using the Back Propagation Through Time (BPTT) algorithm. BPTT algorithm is modified as follows. 1. Evaluate predictability of robot arm/objects using training error of MTRNN. 2. Assign a preference ratio to each object based on predictability. The preference ratio represents the weight of each object to training. Experiments were conducted using an actual robot moving the arm while a human moves his arm in the robot's camera view. The result of the experiment showed that the proposed method presents better training result of robot arm visuomotor dynamics compared to general training with BPTT.

    Original languageEnglish
    Title of host publication2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence, URAI 2014
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    Pages364-367
    Number of pages4
    ISBN (Electronic)9781479953325
    DOIs
    Publication statusPublished - 2014
    Event2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence, URAI 2014 - Kuala Lumpur, Malaysia
    Duration: 2014 Nov 122014 Nov 15

    Other

    Other2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence, URAI 2014
    CountryMalaysia
    CityKuala Lumpur
    Period14/11/1214/11/15

    Fingerprint

    Robots
    Recurrent neural networks
    Backpropagation
    Experiments
    Cameras

    Keywords

    • Cognitive Developmental Robotics
    • Intrinsic Motivation
    • Recurrent Neural Network

    ASJC Scopus subject areas

    • Artificial Intelligence
    • Human-Computer Interaction

    Cite this

    Nishide, S., Nobuta, H., Okuno, H. G., & Ogata, T. (2014). Applying intrinsic motivation for visuomotor learning of robot arm motion. In 2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence, URAI 2014 (pp. 364-367). [7057370] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/URAI.2014.7057370

    Applying intrinsic motivation for visuomotor learning of robot arm motion. / Nishide, Shun; Nobuta, Harumitsu; Okuno, Hiroshi G.; Ogata, Tetsuya.

    2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence, URAI 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 364-367 7057370.

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

    Nishide, S, Nobuta, H, Okuno, HG & Ogata, T 2014, Applying intrinsic motivation for visuomotor learning of robot arm motion. in 2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence, URAI 2014., 7057370, Institute of Electrical and Electronics Engineers Inc., pp. 364-367, 2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence, URAI 2014, Kuala Lumpur, Malaysia, 14/11/12. https://doi.org/10.1109/URAI.2014.7057370
    Nishide S, Nobuta H, Okuno HG, Ogata T. Applying intrinsic motivation for visuomotor learning of robot arm motion. In 2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence, URAI 2014. Institute of Electrical and Electronics Engineers Inc. 2014. p. 364-367. 7057370 https://doi.org/10.1109/URAI.2014.7057370
    Nishide, Shun ; Nobuta, Harumitsu ; Okuno, Hiroshi G. ; Ogata, Tetsuya. / Applying intrinsic motivation for visuomotor learning of robot arm motion. 2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence, URAI 2014. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 364-367
    @inproceedings{09bda1d368ff4284aa6f216fd739ec17,
    title = "Applying intrinsic motivation for visuomotor learning of robot arm motion",
    abstract = "In this paper, we present a method to apply intrinsic motivation for improving visuomotor learning of robot's arm with external object in view. Multiple Timescales Recurrent Neural Network (MTRNN) is utilized for learning the robot arm/external object dynamics. Training of MTRNN is done using the Back Propagation Through Time (BPTT) algorithm. BPTT algorithm is modified as follows. 1. Evaluate predictability of robot arm/objects using training error of MTRNN. 2. Assign a preference ratio to each object based on predictability. The preference ratio represents the weight of each object to training. Experiments were conducted using an actual robot moving the arm while a human moves his arm in the robot's camera view. The result of the experiment showed that the proposed method presents better training result of robot arm visuomotor dynamics compared to general training with BPTT.",
    keywords = "Cognitive Developmental Robotics, Intrinsic Motivation, Recurrent Neural Network",
    author = "Shun Nishide and Harumitsu Nobuta and Okuno, {Hiroshi G.} and Tetsuya Ogata",
    year = "2014",
    doi = "10.1109/URAI.2014.7057370",
    language = "English",
    pages = "364--367",
    booktitle = "2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence, URAI 2014",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    address = "United States",

    }

    TY - GEN

    T1 - Applying intrinsic motivation for visuomotor learning of robot arm motion

    AU - Nishide, Shun

    AU - Nobuta, Harumitsu

    AU - Okuno, Hiroshi G.

    AU - Ogata, Tetsuya

    PY - 2014

    Y1 - 2014

    N2 - In this paper, we present a method to apply intrinsic motivation for improving visuomotor learning of robot's arm with external object in view. Multiple Timescales Recurrent Neural Network (MTRNN) is utilized for learning the robot arm/external object dynamics. Training of MTRNN is done using the Back Propagation Through Time (BPTT) algorithm. BPTT algorithm is modified as follows. 1. Evaluate predictability of robot arm/objects using training error of MTRNN. 2. Assign a preference ratio to each object based on predictability. The preference ratio represents the weight of each object to training. Experiments were conducted using an actual robot moving the arm while a human moves his arm in the robot's camera view. The result of the experiment showed that the proposed method presents better training result of robot arm visuomotor dynamics compared to general training with BPTT.

    AB - In this paper, we present a method to apply intrinsic motivation for improving visuomotor learning of robot's arm with external object in view. Multiple Timescales Recurrent Neural Network (MTRNN) is utilized for learning the robot arm/external object dynamics. Training of MTRNN is done using the Back Propagation Through Time (BPTT) algorithm. BPTT algorithm is modified as follows. 1. Evaluate predictability of robot arm/objects using training error of MTRNN. 2. Assign a preference ratio to each object based on predictability. The preference ratio represents the weight of each object to training. Experiments were conducted using an actual robot moving the arm while a human moves his arm in the robot's camera view. The result of the experiment showed that the proposed method presents better training result of robot arm visuomotor dynamics compared to general training with BPTT.

    KW - Cognitive Developmental Robotics

    KW - Intrinsic Motivation

    KW - Recurrent Neural Network

    UR - http://www.scopus.com/inward/record.url?scp=84988239793&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84988239793&partnerID=8YFLogxK

    U2 - 10.1109/URAI.2014.7057370

    DO - 10.1109/URAI.2014.7057370

    M3 - Conference contribution

    AN - SCOPUS:84988239793

    SP - 364

    EP - 367

    BT - 2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence, URAI 2014

    PB - Institute of Electrical and Electronics Engineers Inc.

    ER -