Thesis: Intelligent Scenario Selection under Simulation Uncertainty for the Validation of Autonomous Systems (m/f/d)

Systematic literature research on methods of scenario selection (Adaptive Sampling, Importance Sampling) and regression processes (Neural Networks, Gaussian Processes). In-depth analysis of modern Uncertainty-Quantification methods focusing on Conformal Prediction. Development of a hybrid approach by combining established methods from simulation, ML and uncertainty modeling. Construction of test rooms with targeted mis-evaluations and definition of suitable evaluation metrics. Analysis of sampling sensitivity and their influence on methodological robustness. Comparison, evaluation and critical analysis of the developed approaches based on simulated critique spaces.

Continuous technical studies with information technology components, e.g. Data Science, Automation technology, physics, electrical engineering or comparable. Basic knowledge of machine learning, deep learning is an advantage. Ideally, first experiences with frameworks such as PyTorch or Scikit-learn. Previous experience with validation methods of advantage. Programming skills in an object-oriented language (preferably Python). Self-employed and systematic work. Communication skills in dealing with colleagues and project participants. Very knowledge of German and English in word and writing.

For us, it is self-evident to offer you optimal framework conditions. These include: Welcomeday and onboarding program Attractive (except) tariff compensation Flexible and family-friendly working hours through sliding time/lifetime account and home office control 30 days of annual leave and special leave under collective agreement Excellent occupational pensions Exciting tasks on innovative products in a growing marine company Grant to the job ticket or Germany ticket Corporate fitness with nationwide partners Bike leasing Comprehensive additional services / attractive external offers Individual learning and development opportunities in presence and digital Comprehensive health management including prevention offers Close cooperation with executives and employees

Collaborative cooperation and respect in dealing with each other – you can find this with us for over 185 years! If this is just as important to you, you can now apply online to specify your salary presentation and your current notice period.

We value diversity

We value diversity and therefore welcome all applications regardless of gender, nationality, ethnic and social origin, religion/worldview, disability, age as well as sexual orientation and identity. .