The recent rapid development of deep-learning-based control strategies has made the reality gap a critical issue at the forefront of robotics, especially for legged robots. We propose a novel system identification framework, Progressive Bayesian Optimisation (ProBO), to bridge the reality gap by tuning simulation parameters. Since dynamic locomotion trajectories are usually harder to narrow the reality gap than their static counterpart, we train a Gaussian process model with the easier trajectory data set and make it a prior to start the learning process of a harder one. We implement ProBO on a quadruped robot to narrow the reality gaps of a set of bounding gaits at different speeds. Results show that our methods can outperform all other alternatives after training the initial gait.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Bridging the Reality Gap via Progressive Bayesian Optimisation


    Weitere Titelangaben:

    Lect. Notes in Networks, Syst.


    Beteiligte:
    Cascalho, José M. (Herausgeber:in) / Tokhi, Mohammad Osman (Herausgeber:in) / Silva, Manuel F. (Herausgeber:in) / Mendes, Armando (Herausgeber:in) / Goher, Khaled (Herausgeber:in) / Funk, Matthias (Herausgeber:in) / Yu, Chen (Autor:in) / Rosendo, Andre (Autor:in)

    Kongress:

    Climbing and Walking Robots Conference ; 2022 ; Ponta Delgada, Portugal September 12, 2022 - September 14, 2022



    Erscheinungsdatum :

    2022-08-25


    Format / Umfang :

    12 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Progressive Bayesian Neural Networks

    Schnaus, Dominik | Deutsches Zentrum für Luft- und Raumfahrt (DLR) | 2021

    Freier Zugriff

    Progressive Subdivision Curves for Aerodynamic Shape Optimisation

    Masters, Dominic A. / Taylor, Nigel J. / Rendall, T. et al. | AIAA | 2016


    Progressive Subdivision Curves for Aerodynamic Shape Optimisation (AIAA 2016-0559)

    Masters, Dominic A. / Taylor, Nigel J. / Rendall, T. et al. | British Library Conference Proceedings | 2016


    Optimisation of passive fire protection for offshore structures using progressive collapse techniques

    Rogers, C. P. / Ramsden, M. | British Library Conference Proceedings | 1995


    Low-noise OGV design for broadband noise using bayesian optimisation

    Paruchuri, Chaitanya C. / Vellanki, Pratibha / Kalyan, Anuroopa et al. | AIAA | 2019