Miniature Autonomy as One Important Testing Means in the Development of Machine Learning Methods for Autonomous Driving: How ML-based Autonomous Driving could be Realized on a 1:87 Scale

Tim Tiedemann, Jonas Fuhrmann, Sebastian Paulsen, Thorben Schnirpel, Nils Schönherr, Bettina Buth, Stephan Pareigis

Abstract

In the current state of autonomous driving machine learning methods are dominating, especially for the environment recognition. For such solutions, the reliability and the robustness is a critical question. A “miniature autonomy” with model vehicles at a small scale could be beneficial for different reasons. Examples are (1) the testability of dangerous and close-to-crash edge cases, (2) the possibility to test potentially dangerous concepts as end-to-end learning or combined inference and learning phases, (3) the need to optimize algorithms thoroughly, and (4) a potential reduction of test mile counts. Presented is the motivation for miniature autonomy and a discussion of testing of machine learning methods. Finally, two currently set up platforms including one with an FPGA-based TPU for ML acceleration are described.

Download


Paper Citation


in Harvard Style

Tiedemann T., Fuhrmann J., Paulsen S., Schnirpel T., Schönherr N., Buth B. and Pareigis S. (2019). Miniature Autonomy as One Important Testing Means in the Development of Machine Learning Methods for Autonomous Driving: How ML-based Autonomous Driving could be Realized on a 1:87 Scale.In Proceedings of the 16th International Conference on Informatics in Control, Automation and Robotics - Volume 2: ICINCO, ISBN 978-989-758-380-3, pages 483-488. DOI: 10.5220/0007955704830488


in Bibtex Style

@conference{icinco19,
author={Tim Tiedemann and Jonas Fuhrmann and Sebastian Paulsen and Thorben Schnirpel and Nils Schönherr and Bettina Buth and Stephan Pareigis},
title={Miniature Autonomy as One Important Testing Means in the Development of Machine Learning Methods for Autonomous Driving: How ML-based Autonomous Driving could be Realized on a 1:87 Scale},
booktitle={Proceedings of the 16th International Conference on Informatics in Control, Automation and Robotics - Volume 2: ICINCO,},
year={2019},
pages={483-488},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0007955704830488},
isbn={978-989-758-380-3},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Conference on Informatics in Control, Automation and Robotics - Volume 2: ICINCO,
TI - Miniature Autonomy as One Important Testing Means in the Development of Machine Learning Methods for Autonomous Driving: How ML-based Autonomous Driving could be Realized on a 1:87 Scale
SN - 978-989-758-380-3
AU - Tiedemann T.
AU - Fuhrmann J.
AU - Paulsen S.
AU - Schnirpel T.
AU - Schönherr N.
AU - Buth B.
AU - Pareigis S.
PY - 2019
SP - 483
EP - 488
DO - 10.5220/0007955704830488