Assembly tasks performed with a robot often fail due to unforeseen situations, regardless of the fact that we carefully learned and optimized the assembly policy. This problem is even more present in humanoid robots acting in an unstructured environment where it is not possible to anticipate all factors that might lead to the failure of the given task. In this work, we propose a concurrent LfD framework, which associates demonstrated exception strategies to the given context. Whenever a failure occurs, the proposed algorithm generalizes past experience regarding the current context and generates an appropriate policy that solves the assembly issue. For this purpose, we applied PCA on force/torque data, which generates low dimensional descriptor of the current context. The proposed framework was validated in a peg-in-hole (PiH) task using Franka-Emika Panda robot. ; This is the author submitted version to ICRA 2020. For the publisher version, please access 10.1109/ICRA40945.2020.9197480.


    Access

    Download


    Export, share and cite



    Title :

    Learning of Exception Strategies in Assembly Tasks


    Contributors:

    Publication date :

    2020-05-31



    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Classification :

    DDC:    629





    Towards Learning by Demonstration for Industrial Assembly Tasks

    Hernandez Moreno, Victor / Carmichael, Marc G. / Deuse, Jochen | Springer Verlag | 2023

    Free access

    TELEOPERATION FOR EXCEPTION HANDLING

    PEDERSEN LIAM / MORTAZAVI ALI / WITWICKI STEFAN et al. | European Patent Office | 2020

    Free access

    Exception handling for autonomous vehicles

    LI DONG / MCNAUGHTON MATTHEW / YEHOSHUA SHIR et al. | European Patent Office | 2022

    Free access

    Porsche 928 S4, l exception

    Porsche,Stuttgart,DE | Automotive engineering | 1987