Jobs bei Cariad in Deutschland
Master Thesis - Development of a Visual SLAM System for Challenging Road Topographies (f / m / d)
Zusammenfassung
An automotive software company in Germany is offering a Master Thesis opportunity focused on developing a robust monocular visual SLAM system. You will engage in benchmarking algorithms and validating them in real-world conditions, while gaining hands-on experience. Ideal candidates are students in a relevant field with programming skills in Python and some knowledge in Computer Vision. This position allows for remote work within Germany and lasts for 6 months.
Qualifikationen
- Students must be enrolled in a relevant field such as Computer Science or Robotics.
- Knowledge of programming in Python and Computer Vision is required.
- Familiarity with SLAM methods and libraries is preferred.
Aufgaben
- Conduct literature review on monocular visual SLAM methods.
- Benchmark algorithms on diverse road topographies.
- Develop new methods for improving SLAM robustness.
- Document and present results in the Master Thesis.
- Prepare results for publication.
Kenntnisse
Ausbildung
We are CARIAD, the automotive software company of the Volkswagen Group. Our teams build automotive software platforms and digital customer functions for iconic brands like Audi, Volkswagen, and Porsche - supporting the Volkswagen Group in becoming the leading automotive technology company. With CARIDIANS in Germany, the USA, China, Estonia, and India, we are transforming automotive mobility for everyone.
Join us and be part of this exciting journey!
Our "Chassis" team is excited to offer a Master Thesis opportunity focused on developing a monocular visual SLAM system that remains robust under challenging road topographies. Unlike conventional SLAM methods that often assume flat, uniform roads, your work will target real-world edge cases such as steep slopes, crests, cobblestones, gravel, and off-road conditions.
This project will involve benchmarking state-of-the-art monocular SLAM algorithms, designing improvements for robustness, and validating them on a unique multi-city driving dataset. You will also gain hands-on experience by evaluating your system in real-world driving scenarios.
Research question: How can monocular SLAM be made robust to diverse and non-flat road topographies in real-world driving?
- Conduct a literature review on state-of-the-art monocular visual SLAM methods
- Benchmark selected algorithms on diverse road topographies and identify failure modes
- Develop and implement new methods to improve SLAM robustness under difficult conditions (e.g., steep slopes, low texture, dynamic traffic)
- Document and present the results in your Master Thesis
- Prepare results for publication in a scientific venue
- Enrolled student in Computer Science, Mechatronics, Electrical Engineering, Robotics, or comparable field
- Knowledge in the field of Computer Vision, and its libraries like OpenCV and NumPy
- Programming skills in Python, previous experience with Tensorflow and / or PyTorch
- Familiarity with SLAM, structure-from-motion, or visual odometry is a strong plus
- Interest in optimization, geometry, and 3D perception
- Team-minded, growth-oriented, communicative, and eager to tackle challenging real-world problems
- Remote work options within Germany
- Duration: 6 months
- 35 h / week
At CARIAD, we embrace individuality and diversity because we believe our differences make us stronger. We actively seek to build teams with a variety of backgrounds, perspectives, and experiences. Our goal is to create an environment where everyone feels valued and empowered to contribute. If you need assistance with your application due to a disability, please reach out to us at careers@cariad.technology - we are happy to support you.
* Der Gehaltsbenchmark wird auf Basis der Zielgehälter bei führenden Unternehmen in der jeweiligen Branche ermittelt und dient Premium-Nutzer:innen als Richtlinie zur Bewertung offener Positionen und als Orientierungshilfe bei Gehaltsverhandlungen. Der Gehaltsbenchmark wird nicht direkt vom Unternehmen angegeben. Er kann deutlich über bzw. unter diesem Wert liegen.