Safely Entering the Deep: A Review of Verification and Validation for Machine Learning and a Challenge Elicitation in the Automotive Industry

Markus Borg; Cristofer Englund; Krzysztof Wnuk; Boris Duran; Christoffer Levandowski; Shenjian Gao; Yanwen Tan; Henrik Kaijser; Henrik Lönn; Jonas Törnqvist

doi:10.2991/jase.d.190131.001

Volume 1, Issue 1, 2020, Pages 1 - 19

Safely Entering the Deep: A Review of Verification and Validation for Machine Learning and a Challenge Elicitation in the Automotive Industry

Downloads:

2,491

Full-Text Views:

89

Citations (Scopus):

31

Citations (Crossref):

0

Download PDF

View Full-Text

Authors

Markus Borg¹^{, *}, Cristofer Englund¹, Krzysztof Wnuk², Boris Duran¹, Christoffer Levandowski³, Shenjian Gao², Yanwen Tan², Henrik Kaijser⁴, Henrik Lönn⁴, Jonas Törnqvist³

¹

RISE Research Institutes of Sweden AB, Scheelevägen 17, SE-223 70 Lund, Sweden

²

Blekinge Institute of Technology, Valhallavägen 1, SE-371 41 Karlskrona, Sweden

³

QRTECH AB, Flöjelbergsgatan 1C, SE-431 35 Mölndal, Sweden

⁴

AB Volvo, Volvo Group Trucks Technology, SE-405 08 Gothenburg, Sweden

^*

Corresponding author. Email: markus.borg@ri.se

Corresponding Author

Markus Borg

Article History

Received 28 May 2018
Accepted 13 December 2018
Available Online 31 January 2019

DOI: https://doi.org/10.2991/jase.d.190131.001
Keywords: Deep learning
Safety-critical systems
Machine learning
Verification and validation
ISO 26262
Abstract: Deep neural networks (DNNs) will emerge as a cornerstone in automotive software engineering. However, developing systems with DNNs introduces novel challenges for safety assessments. This paper reviews the state-of-the-art in verification and validation of safety-critical systems that rely on machine learning. Furthermore, we report from a workshop series on DNNs for perception with automotive experts in Sweden, confirming that ISO 26262 largely contravenes the nature of DNNs. We recommend aerospace-to-automotive knowledge transfer and systems-based safety approaches, for example, safety cage architectures and simulated system test cases.
Open Access: This is an open access article distributed under the CC BY-NC 4.0 license (http://creativecommons.org/licenses/by-nc/4.0/).

Next Article In Issue

Cite This Article

ris

TY  - JOUR
AU  - Markus Borg
AU  - Cristofer Englund
AU  - Krzysztof Wnuk
AU  - Boris Duran
AU  - Christoffer Levandowski
AU  - Shenjian Gao
AU  - Yanwen Tan
AU  - Henrik Kaijser
AU  - Henrik Lönn
AU  - Jonas Törnqvist
PY  - 2019
DA  - 2019/01/31
TI  - Safely Entering the Deep: A Review of Verification and Validation for Machine Learning and a Challenge Elicitation in the Automotive Industry
JO  - Journal of Automotive Software Engineering
SP  - 1
EP  - 19
VL  - 1
IS  - 1
SN  - 2589-2258
UR  - https://doi.org/10.2991/jase.d.190131.001
DO  - https://doi.org/10.2991/jase.d.190131.001
ID  - Borg2019
ER  -

download

copy

enw

bib

Journal Menu

Journal of Automotive Software Engineering

Cite This Article