Bachelor/Semester/Master thesis: A VLM-Based Scenario Analysis of Autonomous Driving

27.02.2025, Abschlussarbeiten, Bachelor- und Masterarbeiten

Background:

The development and validation of autonomous driving systems increasingly rely on scenario-based testing with simulation in the loop and hardware in the loop. Tools such as CommonRoad offer a wide range of realistic traffic scenarios—from urban and rural environments to highways—that incorporate dynamic obstacles, complex road networks, and diverse goal regions. However, conventional simulation methods often fail to capture all safety-critical events that arise from complex interactions among vehicles, environmental factors, and traffic flow.

To address this gap, we propose the use of Vision-Language Models (VLMs). These innovative models combine visual perception with natural language processing to accurately detect and classify safety-relevant events, enabling the systematic identification of both overt and subtle hazardous conditions.

Objective:

The primary objective of this project is to enhance the safety evaluation of autonomous driving systems by employing VLMs to identify safety-critical scenarios within the CommonRoad framework.

We Offer:
An exciting and forward-looking research environment.
The opportunity to work with a state-of-the-art software stack for autonomous driving.
The possibility to publish scientific results (subject to merit).
Flexible supervision and the option to conduct the work in either German or English.

Requirements (What You Should Bring):
Initiative and a creative, problem-solving mindset.
Excellent proficiency in either German or English.
Advanced knowledge of Python or C++.
Experience in machine learning, particularly with Vision-Language Models; familiarity with motion planning is an advantage.
Familiarity with traffic simulation tools such as CommonRoad is desirable.
Proficiency with software development tools such as Git and Ubuntu is advantageous.

Work can begin immediately. If you are interested in this topic, please send an email with a brief cover letter explaining why you are fascinated by this subject, along with a current transcript of records and your resume, to: yuan_avs.gao@tum.de.

Kontakt: yuan_avs.gao@tum.de