SceML: a graphical modeling framework for scenario-based testing of autonomous vehicles

Resource type
Barbara Schütt, Thilo Braun, Stefan Otten, Eric Sax
Ensuring the functional correctness and safety of autonomous ve-hicles is a major challenge for the automotive industry. However,exhaustive physical test drives are not feasible, as billions of drivenkilometers would be required to obtain reliable results. Scenario-based testing is an approach to tackle this problem and reducenecessary test drives by replacing driven kilometers with simula-tions of relevant or interesting scenarios. These scenarios can begenerated or extracted from recorded data with machine learningalgorithms or created by experts. In this paper, we propose a novelgraphical scenario modeling language. The graphical frameworkallows experts to create new scenarios or review ones designed byother experts or generated by machine learning algorithms. Thescenario description is modeled as a graph and based on behaviortrees. It supports different abstraction levels of scenario descriptionduring software and test development. Additionally, the graph-based structure provides modularity and reusable sub-scenarios,an important use case in scenario modeling. A graphical visualiza-tion of the scenario enhances comprehensibility for different users.The presented approach eases the scenario creation process andincreases the usage of scenarios within development and testingprocesses.
Research focus
Safe and Intelligent Vehicles
Download .bib
Download .bib
Published by
Thilo Braun