Scenario-Based Validation of Automated Driving Systems

Abstract

To assess the safety of automated driving systems (ADS), all potentially critical situations have to be considered. One way to do so is to test the function performance in scenarios which lead to these situations. A scenario is a description of an evolution of traffic situations, consisting of snapshots capturing important intermediate states and operators specifying what happens between these states.

The number of relevant scenarios is very large, even if the ADS is supposed to operate in a restricted domain. Therefore, testing must rely heavily on virtual, largely automatized exploration of scenario spaces. For that, classes of scenarios have to be described formally, to enable the dynamic generation of test cases.

The contribution delineates a general approach to safety assessment by virtual testing. It discusses in particular the nature and building blocks of a formal scenario language and the construction of test specifications.

This research was partially funded by the German Federal Ministry for Economic Affairs and Energy, Grant No. 19A15012F (PEGASUS), based on a decision by the Parliament of the Federal Republic of Germany. The responsibility for the content lies with the author.

A previous version of this article (in German) was presented at the AAET 2018 (Automatisiertes und vernetztes Fahren), Brunswick, Germany, 2018.