SUNRISE D3.1 : Report on baseline analysis of existing Methodology: Safety assUraNce fRamework for connected, automated mobIlity SystEms
2023 (English)Report (Other academic)
Abstract [en]
Safety assurance of cooperative, connected, and automated mobility (CCAM) systems is crucial for their successful adoption in society. To demonstrate that such systems are safe in their complete operational design domains (ODDs) requires robust safety argumentation. The aim of the SUNRISE project is to develop and demonstrate a safety assurance framework (SAF) for the test and safety validation of a varied scope of such systems. Scenario-based testing methods is believed to become an important part of the safety assessment approach for automated driving systems (ADSs). The SUNRISE project’s forerunner project HEADSTART developed a methodology for safety validation of connected and automated vehicles centred around scenario-based testing, a methodology that SUNRISE will develop further and integrate as a part of the SUNRISE SAF. Focus for Work Package 3 of the SUNRISE project is to define and condense an overall methodology to support the safety argumentation using data- and knowledge-driven, scenario-based testing. This report presents a literature study and baseline tracking of the existing scenario-based methodologies, especially, based on the knowledge and literature review of the HEADSTART project. First, the SUNRISE SAF and scenario-based methodologies are introduced including a suitable taxonomy. Second, the HEADSTART method is summarized in detail. Third, scenario-based methodologies from other projects are described. Fourth, an overview of relevant standardization efforts is presented with a particular focus on the ISO 3450X series “Road vehicles – Test scenarios for automated driving systems”. Fifth, other initiatives related to scenario-based safety assessment (mainly outside the EU) are described. Sixth, an extensive analysis is presented comparing the HEADSTART methodology with the other described initiatives. Seventh and final, the findings are summarised in the conclusions. The SUNRISE methodology will use the HEADSTART methodology as input complemented with other existing best practices documented in this report. For areas that was in focus for the HEADSTART project, such as scenario concept, test scenario selection and test scenario allocation, the HEADSTART method is concluded to be well defined for future development. Important is that the SUNRISE scenario concept need to be versatile and adoptable for scenario concepts used in all relevant existing scenario databases. As far as possible the scenario concept should also be adoptable for possible future relevant scenario concepts. Other areas, like scenario sources, scenario generation, and scenario databases, were not in focus for HEASTART and only conceptually defined. The SUNRISE data framework is essential to solve these parts as SUNRISE, like HEADSTART, relays on external scenario databases. Further, the HEADSTART methodology needs to be complemented with elements like risk assessment, monitoring in order to identify unknown scenarios, and qualitative and quantitative metrics to determine the completeness of a scenario database.
Place, publisher, year, edition, pages
2023. , p. 107
National Category
Vehicle Engineering
Identifiers
URN: urn:nbn:se:ri:diva-68610OAI: oai:DiVA.org:ri-68610DiVA, id: diva2:1819878
Note
Contributing authors: Oliver Bartels (BASt), Christian Berger, Tayssir Bouraffa, Cagri Kaya (CHAL), Mirko Muro (CRF), Anastasia Bolovinou, Ilias Panangiotopoulos (ICCS), Daniel Becker (ika), Ashfaq Farooqui, Martin Skoglund, Per Stålberg, Fredrik Warg (RISE), Emmanuel Arnoux (RSA), Olaf Op den Camp (TNO), Ghada Ben Nejma (VED), Antonio Bruto da Costa, Patrick Irvine (WMG).
Horizon Research and Innovation Actions | Project No. 101069573. Call HORIZON-CL5-2021-D6-01
2023-12-152023-12-152024-04-24Bibliographically approved