Holistic perspectives on safety of Automated Driving Systems

To join in real life, please sign up here .

ABSTRACT: In recent years, enormous investments in Automated Driving Systems (ADSs) have distinctly advanced ADS technologies. Despite promises made by several high profile auto-makers, it has however become clear that the challenges involved for deploying ADS have been drastically underestimated.Contrary to previous generations of automotive systems, common design, development, verification and validation methods for safety critical systems do not suffice to cope with the increased complexity and operational uncertainties of an ADS. In this talk I will present the key findings from a review paper conducted with the aim to provide an understanding of existing methods for providing safety evidence and, in particular, identifying the associated challenges and gaps pertaining to the use of each method. Four categories of methods: design techniques, verification and validation methods, run-time risk assessment, and run-time (self-)adaptation, are presented and the eight challenges (identified in the paper) help provide insights into a selection of the reviewed methods. Finally, the derived set of research gaps will be discussed.

BIO: Magnus Gyllenhammar pursues a PhD at KTH Royal Institute of Technology as part of his employment at Zenseact, Gothenburg, Sweden. His research focuses on finding efficient strategies for safety argumentation of ADSs, especially focusing on precautionary safety and dynamic risk assessment in relation to the fulfilment of a quantitative risk norm. He received his MSc. in Engineering Physics, major in Complex Adaptive System, from Chalmers University of Technology, in 2016. In 2018, he joined Zenseact (then Zenuity) and has since worked on creating and realising data-driven strategies for verification and safety argumentation of ADSs.