This paper presents a framework that enables a robot to discover various object categories through interaction. The categories are described using action-effect relations, i.e. sensorimotor contingencies rather than more static shape or appearance representation. The framework provides a functionality to classify objects and the resulting categories, associating a class with a specific module. We demonstrate the performance of the framework by studying a pushing behavior in robots, encoding the sensorimotor contingencies and their predictability with Gaussian Processes. We show how entropy-based action selection can improve object classification and how functional categories emerge from the similarities of effects observed among the objects. We also show how a multidimensional action space can be realized by parameterizing pushing using both position and velocity. ; QC 20160422


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    A sensorimotor learning framework for object categorization


    Beteiligte:
    Högman, Virgile (Autor:in) / Björkman, Mårten (Autor:in) / Maki, Atsuto (Autor:in) / Kragic, Danica (Autor:in)

    Erscheinungsdatum :

    2016-01-01


    Anmerkungen:

    ISI:000388682400003



    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Klassifikation :

    DDC:    629



    Learning semantic object parts for object categorization

    Leibe, B. / Ettlin, A. / Schiele, B. | British Library Online Contents | 2008


    Learning Sensorimotor Abstractions

    Cordero Marcos, María Isabel | BASE | 2011

    Freier Zugriff

    A cortical framework for invariant object categorization and recognition

    Rodrigues, J. M. F. / du Buf, J. M. H. | BASE | 2011

    Freier Zugriff

    Reinforcement learning using sensorimotor traces

    Li, Jingxian | BASE | 2013

    Freier Zugriff

    Gaussian Processes for Object Categorization

    Kapoor, A. / Grauman, K. / Urtasun, R. et al. | British Library Online Contents | 2010