Abstract
Deep Neural Networks (DNNs) are one of many supervised machine learning approaches. These data-driven deep learning algorithms are revolutionizing the modern society in domains such as image processing, medicine and automotive. In the field of computer vision, DNNs are outperforming the traditional approaches that use hand-crafted feature extractors. As a result, researchers and developers in the automotive industry are using DNNs for the perception tasks of automated driving. Compared to traditional rule-based approaches, DNNs raise new safety challenges that have to be solved. There are four major building blocks in the development pipeline of DNNs: (1) functionality definition, (2) data set specification, selection and preparation, (3) development and evaluation, and (4) deployment and monitoring. This paper gives an overview of the safety challenges along the whole development pipeline of DNN, proposes potential solutions that are necessary to create safe DNNs and shows first experimental results of DNN performing object detection.
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
Abu Alhaija, H., et al.: Augmented Reality Meets Computer Vision: Efficient Data Generation for Urban Driving Scenes. arXiv preprint: arXiv:1708.01566 (2017)
Chen, X., et al.: Multi-view 3D object detection network for autonomous driving. In: IEEE Conference on Computer Vision and Pattern Recognition (2017)
Cheng, C., et al.: Runtime Monitoring Neuron Activation Patterns (2018)
Eifel, A., et al.: Multimodal deep learning for robust RGB-D object recognition. In: IEEE International Conference on Intelligent Robots and Systems (2015)
Frtunikj, J., Fuerst, S.: Engineering safe machine learning for automated driving systems. In: Safety-Critical Systems Symposium 2019 (2019)
Gal, Y.: Uncertainty in deep learning, Ph.D. dissertation, University of Cambridge (2016)
Girshick, R.: Fast R-CNN. In: IEEE International Conference on Computer Vision (2015)
ISO 26262 Road vehicles: Functional Safety (2018)
ISO/PRF PAS 21448 Road vehicles: Safety of the intended functionality (2018)
Kim, J., et al.: Robust deep multi-modal learning based on gated information fusion network. In: Computer Vision and Pattern Recognition (2018)
Ku, J., et al.: Joint 3D Proposal Generation and Object Detection from View Aggregation. arXiv preprint: arXiv:1712.02294 (2017)
Koopman, P., Wagner, M.: Challenges in autonomous vehicle testing and validation. SAE Int. J. Transp. Saf. 4(1), 15–24 (2016)
Poibrenski, A., Sprenger, J., Müller, C.: Towards a methodology for training with synthetic data on the example of pedestrian detection in a frame-by-frame semantic segmentation task. In: First Workshop on Software Engineering for AI in Autonomous Systems (2018)
Tang, S., Yuan, Y.: Object Detection based on Convolutional Neural Network (2015)
Ulbrich, S., Menzel, T., Reschka, A., et al.: Defining and substantiating the terms scene, situation, and scenario for automated driving. In: IEEE 18th International Conference on Intelligent Transportation Systems (2015)
Zachary, P., et al.: Putting Image Manipulations in Context: Robustness Testing for Safe Perception (2018)
Zendel, O., et al.: How good is my test data? introducing safety analysis for computer vision. Int. J. Comput. Vis. 125(1–3), 95–109 (2017)
Everingham, M., et al.: The PASCAL Visual Object Classes Challenge 2007 Results (2007)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Frtunikj, J. (2019). Practical Experience Report: Engineering Safe Deep Neural Networks for Automated Driving Systems. In: Romanovsky, A., Troubitsyna, E., Bitsch, F. (eds) Computer Safety, Reliability, and Security. SAFECOMP 2019. Lecture Notes in Computer Science(), vol 11698. Springer, Cham. https://doi.org/10.1007/978-3-030-26601-1_16
Download citation
DOI: https://doi.org/10.1007/978-3-030-26601-1_16
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-26600-4
Online ISBN: 978-3-030-26601-1
eBook Packages: Computer ScienceComputer Science (R0)