Abstract
Image segmentation is used to recognize lanes and markings on the road to provide a basis for autonomous driving. On embedded systems in autonomous miniature vehicles with limited resources, timing aspects have to be considered. Optimisations in hardware design and in the architecture of the artificial neural network speed up the inference such that real-time conditions for autonomous driving can be met. Different architectures for artificial neural networks for segmentation are compared. Inference of the segmentation networks is tested on a TPU, an iPhone, and on a Mac with M1 chip. Experiments show, that inference directly on an iPhone is superior to the other two options.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Tiedemann, T., Schwalb, L., Kasten, M., Grotkasten, R., Pareigis, S.: Miniature autonomy as means to find new approaches in reliable autonomous driving AI method design. Front. Neurorobot. 16, 846355 (2022)
Google, “Coral,” (2022). https://coral.ai
Nikolov, I.: Verfahren zur Fahrbahnverfolgung eines autonomen Fahrzeugs mittels Pure Pursuit und Follow-the-carrot. Thesis, University of Applied Sciences Hamburg, B.S (2009)
Riege, D.: “Segmentierung von Straßenmarkierungen durch maschinelles Lernen für die Miniaturautonomie,” B.Sc. Thesis, HAW Hamburg, Department Informatik, Berliner Tor 7, 12 2021
Kasten, M.: “Hardwareplattform für Autonome Straßenfahrzeuge im Maßstab 1:87,” B.Sc. Thesis, HAW Hamburg, Department Informatik, Berliner Tor 7, 10 2021
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation, May 2015. ar**s in deep residual networks, July 2016. ar**v:1603.05027
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., Chen, L.-C.: MobileNetV2: inverted residuals and linear bottlenecks, March 2019. ar**v:1801.04381
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Riege, D., Pareigis, S., Tiedemann, T. (2023). Real-Time Aspects of Image Segmentation of Road Markings in Miniature Autonomy. In: Unger, H., Schaible, M. (eds) Real-time and Autonomous Systems 2022. Real-Time 2022. Lecture Notes in Networks and Systems, vol 674. Springer, Cham. https://doi.org/10.1007/978-3-031-32700-1_11
Download citation
DOI: https://doi.org/10.1007/978-3-031-32700-1_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-32699-8
Online ISBN: 978-3-031-32700-1
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)