TY - THES A1 - Meyer, Matthias T1 - Graph-based generation of OpenSCENARIO-driving-scenarios for virtual validation of automated driving functions N2 - With the prospect of self-driving cars, challenges in the areas of safety and trust, among others, must be addressed. Virtual testing and validation represents an opportunity to address these challenges. Therefore, in the context of this master thesis, the potential of enriching the virtual test of automated driving functions was explored. A virtual test used for the validation of automated driving functions benefits, among other things, from reproducible, fully comprehensive and automatically generated driving situations. In this thesis, a graph-based methodology was developed that fulfills these requirements to provide such driving situations. This methodology is based on the open standard ASAM OpenSCENARIO as well as the graph library QuikGraph. As basis for this, an OpenSCENARIO-Library was developed as part of this thesis. The functionality is demonstrated by means of an example based on the OpenSCENARIOStandard. The driving situations - made available as a so-called scenarios - are generated according to the following graph-based method: - Create OpenSCENARIO-Graph using both the created OpenSCENARIO-Library and QuikGraph - Analyze OpenSCENARIO-Graph using DFS and find the shortest paths using Hoffman and Pavley K-shortest Path algorithm -- Challenges addressed: Cycle resolution and complying to naming conventions within the OpenSCENARIO-Standard - Provide paths through the OpenSCENARIO-Graph through OpenSCENARIO-Scenarios for validation These driving situations can be used to enrich OpenSCENARIO-compatible driving simulators (in this work: esmini) to enable virtual testing and validation of automated driving functions. Thus, this work represents a small contribution to solve the challenges of self-driving cars. Y1 - 2022 UR - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:573-34377 CY - Ingolstadt ER -