Imagine a future where your car not only understands the road but also comprehends the entire driving environment, adapting in real-time to ensure the safest and most efficient journey
With autonomous vehicles and advanced driver assistance systems (ADAS) on the rise, the challenge lies in making these systems smarter and more responsive to the world around them. If you’re passionate about AI and eager to shape the future of driving, this thesis will put you in the driver’s seat of innovation.
Download as PDF
Student Target Groups:
- Students in ICE, CS or Software Engineering.
Thesis Type:
- Master Project / Master Thesis
Goals and Tasks:
- Conduct a comprehensive literature review on multi-modal data encoding and its application in LLMs.
- Develop methods for encoding and integrating sensory data (e.g., speed, symbols, noise levels, gear, map information, GPS) into LLMs.
- Implement and test these methods within a driving assistant prototype, focusing on real time processing and accuracy
- Evaluate the performance of the LLM-based driving assistant in simulated driving environments with various real-world scenarios.
- Present the results and compile the findings in a detailed report
Required competences and knowledge:
- Strong background in machine learning and natural language processing.
- Programming skills in Python.
- Familiarity with deep learning frameworks(e.g., PyTorch or TensorFlow).
Starting date:
Contact: