Bachelor's and Master's degree in Electrical and Information Engineering with a specialization in Information and Automation at the Karlsruhe Institute of Technology (KIT). Practical experience at Siemens Mobility GmbH in Erlangen and BASF SE in Ludwigshafen. Master's thesis on "Development of a Data-Based Decision Support System for the Classification of PCB Defects" at Siemens AG in Karlsruhe.

Mathematical Description of Uncertainties in Environmental Perception for Highly Automated Driving for Consideration in Trajectory Planning

Highly automated driving currently represents one of the major research fields, offering the prospect of fewer accidents, more efficient road traffic, and greater comfort. The prerequisite for both widespread approval and broad acceptance by the public is the certainty that traffic safety is ensured.
In the automation chain, environmental perception is one of the potential sources of errors, from which safety-critical situations can arise. Real physical sensors are subject to error processes that can be further propagated by statistical threshold decisions, for example. Various machine learning-based algorithms for sensor data processing can also cause an incorrect internal representation of the environment. If this internal representation is assumed to be error-free and deterministic during trajectory planning, it does not meet the safety requirements of road traffic. Similarly, the insertion of general safety mechanisms such as virtual safety distances does not address the problem adequately—consider, for instance, a narrow passage that could no longer be navigated.
The aim of my work is the mathematical description of the uncertainties that arise in the course of environmental perception. These situation-dependent uncertainties can then be taken into account when planning trajectories, allowing for the safe verification of the driving maneuver based on the driving situation