This paper explores the use of Path Integral Methods, particularly several variants of the recent Path Integral Policy Improvement (PI 2 ) algorithm in multidimensional movement parametrized policy learning. We rely on Dynamic Movement Primitives (DMPs) to codify discrete and rhythmic trajectories, and apply the PI 2 -CMA and PI BB methods in the learning of optimal policy parameters, according to different cost functions that inherently encode movement objectives. Additionally we merge both of these variants and propose the PI BB -CMA algorithm, comparing all of them with the vanilla version of PI 2 . From the obtained results we conclude that PI BB -CMA surpasses all other methods in terms of convergence speed and iterative final cost, which leads to an increased interest in its application to more complex robotic problems. ; FEDER Funding supported by the Operational Program Competitive Factors - COMPETE andNational Funding supported by the FCT - Portuguese Science Foundation through scholarship UMINHO/BI/40/2012 inserted in project PTDC/EEACRO/100655/2008 and also project FCOMP-01-0124-FEDER-022674


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Path integral learning of multidimensional movement trajectories


    Beteiligte:
    André, João (Autor:in) / Santos, Cristina (Autor:in) / Costa, Lino (Autor:in)

    Erscheinungsdatum :

    2013-01-01



    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Klassifikation :

    DDC:    629



    Multidimensional optimization of trajectories for Earth observation satellites

    Contant, J.-M. / Winkels, H.-M. / Bernabeu, Y.-A. | Elsevier | 1981


    Path-based Multimodal Trajectories Prediction

    Zhao, Ziqi / Duan, Yiping / Tao, Xiaoming | IEEE | 2022




    EXTRACTING PATH GRAPHS FROM VEHICLE TRAJECTORIES

    Ulbrich, Fritz / Rotter, Simon / Goehring, Daniel et al. | British Library Conference Proceedings | 2016