In the joint research project AUTOtech.agil, employees of the Institute of Control Engineering are researching methods and concepts in the context of the safety assurance of automated road vehicles. Due to the complexity of the systems themselves and the open world in which they will operate in the future, residual risk during operation is unavoidable despite the utmost care taken during development. Structured proof of safety, which must be based on consistent and systematic safety considerations, therefore plays a crucial role.
One of the aims of the safety assurance case in the project is to prove the absence of unreasonable risk for the final demonstration with the help of evidence (e.g. test protocols, hazard lists, etc.) from the development process. The resulting safety argumentation is particularly characterized by the involvement of various functional experts. This safety argumentation is modeled semi-formally (Goal Structuring Notation, GSN).
As part of a master's thesis, methods for evaluating uncertainties in the safety argumentation are to be investigated. In particular, one aim is to identify mathematical approaches for capturing and propagating evidence uncertainties through an argumentation according to literature. Furthermore, a requirements-based selection of a method is to be made. The necessary assumptions for the application of this method to an argumentation have to be defined. Subsequently, the connection (programming as a feature) of the uncertainty elicitation and calculation to a visual GSN editor is to be carried out. A minimal example (partial argumentation) should be used to create a proof of concept, for the implementation of which experts from the project can be consulted in the course of the initially necessary elicitation of evidence uncertainties.
Required are...
What is helpful is experience/knowledge in the context of...