We present a perception system for mobile manipulation tasks. The primary design goal of the proposed system is to minimize human interaction during system setup which is achieved by several means, such as automatic training data generation, the use of simulated training data, and 3D model based geometric matching. We employ a state-of-the art deep-learning based bounding box detector for rough localization of objects and a Point Pair Feature based matching algorithm for 6DoF pose estimation. The proposed approach shows promising results on our recently published dataset for industrial object detection and pose estimation. Furthermore, the system’s performance during four days of live operation at the Automatica 2018 trade fair is analyzed and failure cases are presented and discussed.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    6DoF Pose Estimation for Industrial Manipulation Based on Synthetic Data


    Additional title:

    Springer Proceedings in Advanced Robotics



    Conference:

    International Symposium on Experimental Robotics ; 2018 ; Buenos Aires, Argentina November 05, 2018 - November 08, 2018



    Publication date :

    2020-01-23


    Size :

    10 pages





    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    English




    6DoF Pose Estimation for Industrial Manipulation Based on Synthetic Data

    Brucker, Manuel / Durner, Maximilian / Márton, Zoltán-Csaba et al. | TIBKAT | 2020


    6DoF Vehicle Pose Estimation Using Segmentation-Based Part Correspondences

    Barowski, Thomas / Szczot, Magdalena / Houben, Sebastian | IEEE | 2019


    Ground Plane Polling for 6DoF Pose Estimation of Objects on the Road

    Rangesh, Akshay / Trivedi, Mohan Manubhai | IEEE | 2020


    6DOF Metrology-integrated Robot Control

    Kihlman, Henrik / Loser, Raimund | SAE Technical Papers | 2003


    Transfer Learning for Driver Pose Estimation from Synthetic Data

    Sagmeister, Daniel / Schorkhuber, Dominik / Nezveda, Matej et al. | IEEE | 2023