Ergo, SMIRK is safe : a safety case for a machine learning component in a pedestrian automatic emergency brake system
(2023) In Software Quality Journal 31(2). p.335-403- Abstract
Integration of machine learning (ML) components in critical applications introduces novel challenges for software certification and verification. New safety standards and technical guidelines are under development to support the safety of ML-based systems, e.g., ISO 21448 SOTIF for the automotive domain and the Assurance of Machine Learning for use in Autonomous Systems (AMLAS) framework. SOTIF and AMLAS provide high-level guidance but the details must be chiseled out for each specific case. We initiated a research project with the goal to demonstrate a complete safety case for an ML component in an open automotive system. This paper reports results from an industry-academia collaboration on safety assurance of SMIRK, an ML-based... (More)
Integration of machine learning (ML) components in critical applications introduces novel challenges for software certification and verification. New safety standards and technical guidelines are under development to support the safety of ML-based systems, e.g., ISO 21448 SOTIF for the automotive domain and the Assurance of Machine Learning for use in Autonomous Systems (AMLAS) framework. SOTIF and AMLAS provide high-level guidance but the details must be chiseled out for each specific case. We initiated a research project with the goal to demonstrate a complete safety case for an ML component in an open automotive system. This paper reports results from an industry-academia collaboration on safety assurance of SMIRK, an ML-based pedestrian automatic emergency braking demonstrator running in an industry-grade simulator. We demonstrate an application of AMLAS on SMIRK for a minimalistic operational design domain, i.e., we share a complete safety case for its integrated ML-based component. Finally, we report lessons learned and provide both SMIRK and the safety case under an open-source license for the research community to reuse.
(Less)
- author
- Borg, Markus LU ; Henriksson, Jens ; Socha, Kasper ; Lennartsson, Olof ; Sonnsjö Lönegren, Elias ; Bui, Thanh ; Tomaszewski, Piotr ; Sathyamoorthy, Sankar Raman ; Brink, Sebastian and Helali Moghadam, Mahshid
- organization
- publishing date
- 2023
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Automotive demonstrator, Machine learning safety, Safety case, Safety standards
- in
- Software Quality Journal
- volume
- 31
- issue
- 2
- pages
- 335 - 403
- publisher
- Springer
- external identifiers
-
- scopus:85149021250
- ISSN
- 0963-9314
- DOI
- 10.1007/s11219-022-09613-1
- language
- English
- LU publication?
- yes
- id
- c6413675-622a-4634-919c-ee349cb346f4
- date added to LUP
- 2023-03-16 14:46:21
- date last changed
- 2023-11-21 02:13:30
@article{c6413675-622a-4634-919c-ee349cb346f4, abstract = {{<p>Integration of machine learning (ML) components in critical applications introduces novel challenges for software certification and verification. New safety standards and technical guidelines are under development to support the safety of ML-based systems, e.g., ISO 21448 SOTIF for the automotive domain and the Assurance of Machine Learning for use in Autonomous Systems (AMLAS) framework. SOTIF and AMLAS provide high-level guidance but the details must be chiseled out for each specific case. We initiated a research project with the goal to demonstrate a complete safety case for an ML component in an open automotive system. This paper reports results from an industry-academia collaboration on safety assurance of SMIRK, an ML-based pedestrian automatic emergency braking demonstrator running in an industry-grade simulator. We demonstrate an application of AMLAS on SMIRK for a minimalistic operational design domain, i.e., we share a complete safety case for its integrated ML-based component. Finally, we report lessons learned and provide both SMIRK and the safety case under an open-source license for the research community to reuse.</p>}}, author = {{Borg, Markus and Henriksson, Jens and Socha, Kasper and Lennartsson, Olof and Sonnsjö Lönegren, Elias and Bui, Thanh and Tomaszewski, Piotr and Sathyamoorthy, Sankar Raman and Brink, Sebastian and Helali Moghadam, Mahshid}}, issn = {{0963-9314}}, keywords = {{Automotive demonstrator; Machine learning safety; Safety case; Safety standards}}, language = {{eng}}, number = {{2}}, pages = {{335--403}}, publisher = {{Springer}}, series = {{Software Quality Journal}}, title = {{Ergo, SMIRK is safe : a safety case for a machine learning component in a pedestrian automatic emergency brake system}}, url = {{http://dx.doi.org/10.1007/s11219-022-09613-1}}, doi = {{10.1007/s11219-022-09613-1}}, volume = {{31}}, year = {{2023}}, }