SMIRK: A machine learning-based pedestrian automatic emergency braking system with a complete safety case
2022 (engelsk)Inngår i: Software Impacts, ISSN 2665-9638, Vol. 13, artikkel-id 100352Artikkel i tidsskrift (Fagfellevurdert) Published
Abstract [en]
SMIRK is a pedestrian automatic emergency braking system that facilitates research on safety-critical systems embedding machine learning components. As a fully transparent driver-assistance system, SMIRK can support future research on trustworthy AI systems, e.g., verification & validation, requirements engineering, and testing. SMIRK is implemented for the simulator ESI Pro-SiVIC with core components including a radar sensor, a mono camera, a YOLOv5 model, and an anomaly detector. ISO/PAS 21448 SOTIF guided the development, and we present a complete safety case for a restricted ODD using the AMLAS methodology. Finally, all training data used to train the perception system is publicly available. © 2022 The Authors
sted, utgiver, år, opplag, sider
Elsevier B.V. , 2022. Vol. 13, artikkel-id 100352
Emneord [en]
Advanced driver-assistance system, Automotive demonstrator, Computer vision, Machine learning, Pedestrian automatic emergency braking, Safety case
HSV kategori
Identifikatorer
URN: urn:nbn:se:ri:diva-59901DOI: 10.1016/j.simpa.2022.100352Scopus ID: 2-s2.0-85134628533OAI: oai:DiVA.org:ri-59901DiVA, id: diva2:1686870
Merknad
Funding details: 2019-05871; Funding details: Fellowships Fund Incorporated, FFI; Funding details: VINNOVA; Funding details: Knut och Alice Wallenbergs Stiftelse; Funding text 1: Our thanks go to everyone in the SMILE III project who helped with developing SMIRK’s corresponding safety case. In particular we acknowledge Thanh Bui, Olof Lennartsson, Elias Sonnsjö Lönegren and Sankar Raman Sathyamoorthy. Furthermore, we thank François-Xavier Jegeden for providing advanced technical ESI Pro-SiVIC support. This work was carried out within the SMILE III project financed by Vinnova, FFI, Fordonsstrategisk forskning och innovation under the grant number 2019-05871 and partially supported by the Wallenberg AI, Autonomous Systems and Software Program (WASP) funded by Knut and Alice Wallenberg Foundation, Sweden . Finally, we thank AI Sweden for showcasing the SMIRK data set and helping interested users to download it.
2022-08-112022-08-112025-09-23bibliografisk kontrollert