zurück

AI-based tire type (summer/winter) classification using acoustic signals

Masterthesis

Tire choice affects safety, efficiency, and noise. Today, fleets and inspection teams still rely on manual checks to verify seasonal tire compliance. Your thesis will explore whether passive acoustics—the sound of tire–road interaction captured by microphones—can reliably identify winter vs. summer tires in real traffic.

Design, implement, and evaluate an end-to-end classification pipeline that determines the mounted tire type from audio recorded during normal driving. The solution should be robust to vehicle speed, road surface, and environmental noise.

Your Tasks:

  • Plan and install microphones and measurement equipment in a test vehicle
  • Collect synchronized vehicle and acoustic data for multiple tire types
  • Signal preprocessing and data analysis
  • Model development and training
  • Model performance evaluation

This is what you bring:

  • Independence, motivation, and initiative
  • Good / very good academic record
  • Curiosity and creativity for a challenging, applied topic
  • Nice to have: Experience with Python and AI/ML (e.g., PyTorch/TensorFlow, scikit-learn), signal processing, or acoustics.

This is what we offer:

  • An exciting, industry-relevant topic with real-world impact
  • Close supervision by a competent and motivated team
  • A modern workplace and access to measurement hardware
  • Collaboration with a startup based in Darmstadt

Interesse geweckt? Dann melde dich gerne mit deinen Noten bei