Your Mission:

Understanding ride quality and suspension performance is a crucial part of race car development. The Vehicle Science department in our Motorsports division is dedicated to developing and applying cutting-edge methods in order to improve the performance of our cars and teams. In this thesis, you will focus on the application of driver models based on reinforcement learning, to carry out data driven setup investigations. Examples would be use cases for ride optimization, improving the existing workflow for suspension setup, or control system tuning.  By bridging vehicle simulation models and data-driven digital twins, your work will contribute to a deeper understanding of race car setup sensitivities, vehicle dynamics, and performance in the context of racing. This topic offers the opportunity to apply cutting-edge, scientifically relevant methods in a competitive environment.

Content of the Thesis:

• Investigation of novel reinforcement learning driver models for ride and suspension-related use cases
• Extension of the existing workflow for use case analysis with machine learning methods
• Analysis of the interaction between suspension setup changes and lap time / driver performance, developing novel data driven analysis methods
• Proposals for improvements to the evaluation pipeline, focusing on robustness and interpretability
• Validation of results by comparison against engineering judgement and existing methods

• Ongoing master studies in Computer Science, Mechanical Engineering, Mechatronics, Data Science, Control Engineering or related fields
• Knowledge in the field of vehicle dynamics
• Experience with reinforcement learning and ML frameworks (TensorFlow, PyTorch)
• Good programming skills in Python or similar
• Analytical mindset with interest in linking physical models and ML approaches
• Passion for motorsport and willingness to work in a highly dynamic environment