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

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 machin...

Full description

Saved in:
Bibliographic Details
Published inJournal of Automotive Software Engineering Vol. 1; no. 1; pp. 1 - 19
Main Authors Borg, Markus, Englund, Cristofer, Wnuk, Krzysztof, Duran, Boris, Levandowski, Christoffer, Gao, Shenjian, Tan, Yanwen, Kaijser, Henrik, Lönn, Henrik, Törnqvist, Jonas
Format Journal Article
LanguageEnglish
Published Dordrecht Springer Netherlands 2019
Subjects
Deep learning
ISO 26262
Safety-critical systems
Verification and validation
Machine learning
Online AccessGet full text

Cover

More Information
Summary: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.
ISSN:2589-2258
DOI:10.2991/jase.d.190131.001