Interactive Learning in Human-Robot Collaboration

Tetsuya Ogata, Noritaka Masago, Shigeki Sugano, Jun Tani

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

    6 Citations (Scopus)

    Abstract

    In this paper, we investigated interactive learning between human subjects and robot experimentally, and its essential characteristics are examined using the dynamical systems approach. Our research concentrated on the navigation system of a specially developed humanoid robot called Robovie and seven human subjects whose eyes were covered, making them dependent on the robot for directions. We compared the usual feed-forward neural net-work (FFNN) without recursive connections and the recurrent neural network (RNN). Although the performances obtained with both the RNN and the FFNN improved in the early stages of learning, as the subject changed the operation by learning on its own, all performances gradually became unstable and failed. Results of a questionnaire given to the subjects confirmed that the FFNN gives better mental impressions, especially from the aspect of operability. When the robot used a consolidation-learning algorithm using the rehearsal outputs of the RNN, the performance improved even when interactive learning continued for a long time. The questionnaire results then also confirmed that the subject's mental impressions of the RNN improved significantly. The dynamical systems analysis of RNNs support these differences.

    Original languageEnglish
    Title of host publicationIEEE International Conference on Intelligent Robots and Systems
    Pages162-167
    Number of pages6
    Volume1
    Publication statusPublished - 2003
    Event2003 IEEE/RSJ International Conference on Intelligent Robots and Systems - Las Vegas, NV
    Duration: 2003 Oct 272003 Oct 31

    Other

    Other2003 IEEE/RSJ International Conference on Intelligent Robots and Systems
    CityLas Vegas, NV
    Period03/10/2703/10/31

    Fingerprint

    Recurrent neural networks
    Robots
    Neural networks
    Dynamical systems
    Navigation systems
    Consolidation
    Learning algorithms
    Systems analysis

    ASJC Scopus subject areas

    • Control and Systems Engineering

    Cite this

    Ogata, T., Masago, N., Sugano, S., & Tani, J. (2003). Interactive Learning in Human-Robot Collaboration. In IEEE International Conference on Intelligent Robots and Systems (Vol. 1, pp. 162-167)

    Interactive Learning in Human-Robot Collaboration. / Ogata, Tetsuya; Masago, Noritaka; Sugano, Shigeki; Tani, Jun.

    IEEE International Conference on Intelligent Robots and Systems. Vol. 1 2003. p. 162-167.

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

    Ogata, T, Masago, N, Sugano, S & Tani, J 2003, Interactive Learning in Human-Robot Collaboration. in IEEE International Conference on Intelligent Robots and Systems. vol. 1, pp. 162-167, 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems, Las Vegas, NV, 03/10/27.
    Ogata T, Masago N, Sugano S, Tani J. Interactive Learning in Human-Robot Collaboration. In IEEE International Conference on Intelligent Robots and Systems. Vol. 1. 2003. p. 162-167
    Ogata, Tetsuya ; Masago, Noritaka ; Sugano, Shigeki ; Tani, Jun. / Interactive Learning in Human-Robot Collaboration. IEEE International Conference on Intelligent Robots and Systems. Vol. 1 2003. pp. 162-167
    @inproceedings{e1a4b39f9d3d4e2e9ad6b11bfa7c2f0e,
    title = "Interactive Learning in Human-Robot Collaboration",
    abstract = "In this paper, we investigated interactive learning between human subjects and robot experimentally, and its essential characteristics are examined using the dynamical systems approach. Our research concentrated on the navigation system of a specially developed humanoid robot called Robovie and seven human subjects whose eyes were covered, making them dependent on the robot for directions. We compared the usual feed-forward neural net-work (FFNN) without recursive connections and the recurrent neural network (RNN). Although the performances obtained with both the RNN and the FFNN improved in the early stages of learning, as the subject changed the operation by learning on its own, all performances gradually became unstable and failed. Results of a questionnaire given to the subjects confirmed that the FFNN gives better mental impressions, especially from the aspect of operability. When the robot used a consolidation-learning algorithm using the rehearsal outputs of the RNN, the performance improved even when interactive learning continued for a long time. The questionnaire results then also confirmed that the subject's mental impressions of the RNN improved significantly. The dynamical systems analysis of RNNs support these differences.",
    author = "Tetsuya Ogata and Noritaka Masago and Shigeki Sugano and Jun Tani",
    year = "2003",
    language = "English",
    volume = "1",
    pages = "162--167",
    booktitle = "IEEE International Conference on Intelligent Robots and Systems",

    }

    TY - GEN

    T1 - Interactive Learning in Human-Robot Collaboration

    AU - Ogata, Tetsuya

    AU - Masago, Noritaka

    AU - Sugano, Shigeki

    AU - Tani, Jun

    PY - 2003

    Y1 - 2003

    N2 - In this paper, we investigated interactive learning between human subjects and robot experimentally, and its essential characteristics are examined using the dynamical systems approach. Our research concentrated on the navigation system of a specially developed humanoid robot called Robovie and seven human subjects whose eyes were covered, making them dependent on the robot for directions. We compared the usual feed-forward neural net-work (FFNN) without recursive connections and the recurrent neural network (RNN). Although the performances obtained with both the RNN and the FFNN improved in the early stages of learning, as the subject changed the operation by learning on its own, all performances gradually became unstable and failed. Results of a questionnaire given to the subjects confirmed that the FFNN gives better mental impressions, especially from the aspect of operability. When the robot used a consolidation-learning algorithm using the rehearsal outputs of the RNN, the performance improved even when interactive learning continued for a long time. The questionnaire results then also confirmed that the subject's mental impressions of the RNN improved significantly. The dynamical systems analysis of RNNs support these differences.

    AB - In this paper, we investigated interactive learning between human subjects and robot experimentally, and its essential characteristics are examined using the dynamical systems approach. Our research concentrated on the navigation system of a specially developed humanoid robot called Robovie and seven human subjects whose eyes were covered, making them dependent on the robot for directions. We compared the usual feed-forward neural net-work (FFNN) without recursive connections and the recurrent neural network (RNN). Although the performances obtained with both the RNN and the FFNN improved in the early stages of learning, as the subject changed the operation by learning on its own, all performances gradually became unstable and failed. Results of a questionnaire given to the subjects confirmed that the FFNN gives better mental impressions, especially from the aspect of operability. When the robot used a consolidation-learning algorithm using the rehearsal outputs of the RNN, the performance improved even when interactive learning continued for a long time. The questionnaire results then also confirmed that the subject's mental impressions of the RNN improved significantly. The dynamical systems analysis of RNNs support these differences.

    UR - http://www.scopus.com/inward/record.url?scp=0346780205&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=0346780205&partnerID=8YFLogxK

    M3 - Conference contribution

    AN - SCOPUS:0346780205

    VL - 1

    SP - 162

    EP - 167

    BT - IEEE International Conference on Intelligent Robots and Systems

    ER -