A taxonomy for quality in simulation-based development and testing of automated driving systems

Ensuring safety and performance requirements of automated driving systems is a major challenge for the automotive industry. One way to tackle this challenge is a simulationbased approach. However, to rely on results generated by using this approach, the simulation needs to fulfill certain quality criteria depending on the intended usage. Hence, quality must be measured and determined at many different levels and areas of the testing and developing landscape, providing information with varying abstraction degrees. Additionally, quality not only has to be assessed for the automated driving system but also for the simulation models that simulate the vehicles components. Only if the simulation models have a certain quality level can they be used for the simulation. The taxonomy presented in this paper provides a better understanding of the concept of quality during the development and test process. It introduces the possibility to systematically evaluate whether development steps in this process need to be repeated or further assessed.