Presentado al IEEE International Conference on Robotics and Automation celebrado en Karlsruhe (Alemania) del 6 al 10 de mayo de 2013. ; Dynamic Motor Primitives (DMP) are nowadays widely used as movement parametrization for learning trajectories, because of their linearity in the parameters, rescalation robustness and continuity. However, when learning a movement with DMP, where a set of gaussians distributed along the trajectory is used to approximate an acceleration excitation function, a very large number of gaussian approximations need to be performed. Adding them up for all joints yields too many parameters to be explored, thus requiring a prohibitive number of experiments/simulations to converge to a solution with an optimal (locally or globally) reward. We propose here two strategies to reduce this dimensionality: the first is to explore only the most significant directions in the parameter space, and the second is to add a reduced second set of gaussians that should only optimize the trajectory after fixing the gaussians that approximate the demonstrated movement. ; This work is partially funded by EU Project IntellAct (FP7-269959) and by the Spanish Ministry of Science and Innovation under project PAU+DPI2011-27510. ; Peer Reviewed


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Handling high parameter dimensionality in reinforcement learning with dynamic motor primitives


    Beteiligte:
    Colomé, Adrià (Autor:in) / Alenyà, Guillem (Autor:in) / Torras, Carme (Autor:in)

    Erscheinungsdatum :

    2013-01-01


    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch


    Klassifikation :

    DDC:    629



    Handling high parameter dimensionality in reinforcement learning with dynamic motor primitives

    Colomé Figueras, Adrià / Alenyà Ribas, Guillem / Torras, Carme | BASE | 2013

    Freier Zugriff



    Reinforcement learning to adjust parametrized motor primitives to new situations

    Kober, J. / Wilhelm, A. / Öztop, Erhan et al. | BASE | 2012

    Freier Zugriff