Thesis on AI-Driven Map Matching and Path Prediction on Semantically Enriched Road Networks
SOME IT WORKS. SOME CHANGES WHAT'S POSSIBLE.
SHARE YOUR PASSION.
More than 90% of automotive innovations are based on electronics and software. That's why creative freedom and lateral thinking are so important in the pursuit of truly novel solutions. That’s why our experts will treat you as part of the team from day one, encourage you to bring your own ideas to the table – and give you the opportunity to really show what you can do.
Map matching and path prediction are core capabilities for autonomous driving. Our team at the BMW Group explores data-driven approaches that combine symbolic reasoning and machine learning, operating on real-world map data to improve robustness, accuracy, and interpretability in complex urban environments.
What awaits you?
- You will support modeling road connectivity and constraints in RDF and implementing rule sets to compute the most probable path using a rule-based reasoner.
- Furthermore, you help build features from observations and the road graph, learning graph embeddings and training models to predict the next link or path.
- In addition, you contribute to formulating path prediction as a reinforcement learning problem, integrating graph embeddings and training agents such as DQN or actor-critic.
- Moreover, you will assist in developing sequence-to-sequence or transformer-based models to align GPS trajectories to graph-aligned edge sequences and comparing them to baselines.
- Darüber hinaus wirkst du mit beim Experimentieren mit graph-aware attention or constrained decoding to inject RDF structure and semantics into learning models.
- Additionally, you support designing metrics and scenarios, measuring accuracy, robustness, efficiency, and interpretability, and running ablations on embeddings and semantic attributes.
What should you bring along?
- Studies in computer science, data science, electrical engineering or a related field.
- Solid background in machine learning, including supervised learning and basic reinforcement learning.
- Experience with Python and machine learning or deep learning frameworks such as PyTorch or TensorFlow, plus data processing libraries.
- Familiarity with graph representations and knowledge graphs such as RDF and SPARQL; comfort with graph or knowledge graph embeddings is a plus.
- Understanding of graph neural networks or representation learning methods.
- Software engineering skills for reproducible experiments, including version control, clean code, and benchmarking.
- Good German and English skills.
You are enthused by new technologies and an innovative environment? Then Apply now!
What do we offer?
- Comprehensive mentoring & onboarding.
- Personal & professional development.
- Flexible working hours.
- Mobile work.
- Attractive & fair compensation.
- Student apartments (subject to availability & only at the Munich location).
- And much more, see bmw.jobs/whatweoffer
Start date: earliest start 05/15/2026
Duration: 6 months
Working hours: Full-time
You can find helpful tips on your application and the application process here.
At the BMW Group, we place great importance on equal treatment and equal opportunities. Our recruiting decisions are based on the personality, experience, and skills of the applicants. Learn more here.