Modeling perception errors of automated vehicles requires reference data, but common reference measurement methods either cannot capture uninstructed road users or suffer from vehicle-vehicle-occlusions. Therefore, we propose a method based on a camera-equipped drone hovering over the field of view of the perception system that is to be modeled. From recordings of this advantageous perspective, computer vision algorithms extract object tracks suited as reference. As a proof of concept of our approach, we create and analyze a phenomenological error model of a lidar-based sensor system. From eight hours of simultaneous traffic recordings at an intersection, we extract synchronized state vectors of associated true-positive vehicle tracks. We model the deviations of the full lidar state vectors from the reference as multivariate Gaussians. The dependency of their covariance matrices and mean vectors on the reference state vector is modeled by a fully-connected neural network. By customizing the network training procedure and losses, we are able to achieve consistent results even in sparsely populated areas of the state space. Finally, we show that time dependencies of errors can be considered separately during sampling by an autoregressive model.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Using Drones as Reference Sensors for Neural-Networks-Based Modeling of Automotive Perception Errors**The research leading to these results is funded by the Federal Ministry for Economic Affairs and Energy within the project “VVM - Verification and Validation Methods for Automated Vehicles Level 4 and 5”. The authors would like to thank the consortium for the successful cooperation.


    Beteiligte:
    Krajewski, Robert (Autor:in) / Ross, Michael (Autor:in) / Meister, Adrian (Autor:in) / Thomsen, Fabian (Autor:in) / Bock, Julian (Autor:in) / Eckstein, Lutz (Autor:in)


    Erscheinungsdatum :

    2020-10-19


    Format / Umfang :

    1855328 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Modeling Perception Errors of Automated Vehicles

    Sigl, Martin / Schutz, Christoph / Wagner, Sebastian et al. | IEEE | 2021


    Validation and Verification of Automated Road Vehicles

    Agaram, Venkatesh / Barickman, Frank / Fahrenkrog, Felix et al. | Springer Verlag | 2016


    Automated Emergency Landing System for Drones: SafeEYE Project

    Bektash, O. / Pedersen, Jacob Naundrup / Ramirez Gomez, Aitor et al. | IEEE | 2020


    Automated Vehicles, Urban Robots and Drones: Three Elements of the Automated City

    Loke, Seng W. / Rakotonirainy, Andry | Springer Verlag | 2021