Open-end human-robot interaction from the dynamical systems perspective: Mutual adaptation and incremental learning

    Research output: Contribution to journalArticle

    19 Citations (Scopus)

    Abstract

    In this paper, we experimentally investigated the open-end interaction generated by the mutual adaptation between humans and robot. Its essential characteristic, incremental learning, is 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 used the usual feed-forward neural network (FFNN) without recursive connections and the recurrent neural network (RNN) for the robot control. 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. Next, we used a 'consolidation-learning algorithm' as a model of the hippocampus in the brain. In this method, the RNN was trained by both new data and the rehearsal outputs of the RNN not to damage the contents of current memory. The proposed method enabled the robot to improve performance even when learning continued for a long time (openend). The dynamical systems analysis of RNNs supports these differences and also showed that the collaboration scheme was developed dynamically along with succeeding phase transitions.

    Original languageEnglish
    Pages (from-to)651-670
    Number of pages20
    JournalAdvanced Robotics
    Volume19
    Issue number6
    DOIs
    Publication statusPublished - 2005

    Fingerprint

    Human robot interaction
    Recurrent neural networks
    Dynamical systems
    Robots
    Feedforward neural networks
    Navigation systems
    Consolidation
    Learning algorithms
    Brain
    Phase transitions
    Systems analysis
    Data storage equipment

    Keywords

    • Consolidation learning
    • Dynamical systems
    • Incremantal learning
    • Mutual adaptation
    • Open-end human-robot interaction
    • Recurrent neural network

    ASJC Scopus subject areas

    • Control and Systems Engineering

    Cite this

    @article{0425b66c488b4cd0ad25cb3e672d49c0,
    title = "Open-end human-robot interaction from the dynamical systems perspective: Mutual adaptation and incremental learning",
    abstract = "In this paper, we experimentally investigated the open-end interaction generated by the mutual adaptation between humans and robot. Its essential characteristic, incremental learning, is 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 used the usual feed-forward neural network (FFNN) without recursive connections and the recurrent neural network (RNN) for the robot control. 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. Next, we used a 'consolidation-learning algorithm' as a model of the hippocampus in the brain. In this method, the RNN was trained by both new data and the rehearsal outputs of the RNN not to damage the contents of current memory. The proposed method enabled the robot to improve performance even when learning continued for a long time (openend). The dynamical systems analysis of RNNs supports these differences and also showed that the collaboration scheme was developed dynamically along with succeeding phase transitions.",
    keywords = "Consolidation learning, Dynamical systems, Incremantal learning, Mutual adaptation, Open-end human-robot interaction, Recurrent neural network",
    author = "Tetsuya Ogata and Shigeki Sugano and Jun Tani",
    year = "2005",
    doi = "10.1163/1568553054255655",
    language = "English",
    volume = "19",
    pages = "651--670",
    journal = "Advanced Robotics",
    issn = "0169-1864",
    publisher = "Taylor and Francis Ltd.",
    number = "6",

    }

    TY - JOUR

    T1 - Open-end human-robot interaction from the dynamical systems perspective

    T2 - Mutual adaptation and incremental learning

    AU - Ogata, Tetsuya

    AU - Sugano, Shigeki

    AU - Tani, Jun

    PY - 2005

    Y1 - 2005

    N2 - In this paper, we experimentally investigated the open-end interaction generated by the mutual adaptation between humans and robot. Its essential characteristic, incremental learning, is 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 used the usual feed-forward neural network (FFNN) without recursive connections and the recurrent neural network (RNN) for the robot control. 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. Next, we used a 'consolidation-learning algorithm' as a model of the hippocampus in the brain. In this method, the RNN was trained by both new data and the rehearsal outputs of the RNN not to damage the contents of current memory. The proposed method enabled the robot to improve performance even when learning continued for a long time (openend). The dynamical systems analysis of RNNs supports these differences and also showed that the collaboration scheme was developed dynamically along with succeeding phase transitions.

    AB - In this paper, we experimentally investigated the open-end interaction generated by the mutual adaptation between humans and robot. Its essential characteristic, incremental learning, is 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 used the usual feed-forward neural network (FFNN) without recursive connections and the recurrent neural network (RNN) for the robot control. 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. Next, we used a 'consolidation-learning algorithm' as a model of the hippocampus in the brain. In this method, the RNN was trained by both new data and the rehearsal outputs of the RNN not to damage the contents of current memory. The proposed method enabled the robot to improve performance even when learning continued for a long time (openend). The dynamical systems analysis of RNNs supports these differences and also showed that the collaboration scheme was developed dynamically along with succeeding phase transitions.

    KW - Consolidation learning

    KW - Dynamical systems

    KW - Incremantal learning

    KW - Mutual adaptation

    KW - Open-end human-robot interaction

    KW - Recurrent neural network

    UR - http://www.scopus.com/inward/record.url?scp=22744441536&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=22744441536&partnerID=8YFLogxK

    U2 - 10.1163/1568553054255655

    DO - 10.1163/1568553054255655

    M3 - Article

    AN - SCOPUS:22744441536

    VL - 19

    SP - 651

    EP - 670

    JO - Advanced Robotics

    JF - Advanced Robotics

    SN - 0169-1864

    IS - 6

    ER -