Autonomous Racing: Trajectory Planning Evaluation

Institut

Lehrstuhl für Fahrzeugtechnik

Typ

Semesterarbeit / Masterarbeit /

Inhalt

theoretisch /  

Beschreibung

TUM Autonomous Motorsport is at the forefront of autonomous racing, competing with real open-wheel formula
cars against the best teams in the world. With victories in prestigious competitions such as the Indy Autonomous
Challenge (IAC) and Abu Dhabi Autonomous Racing League (A2RL) — events featuring multi-million-
dollar prize pools — our team has proven itself as a leader in high-speed autonomous driving. These competitions
push the limits of AI-driven motion planning, perception, and control, requiring cutting-edge software capable of
racing at extreme speeds while ensuring safety and reliability. This thesis offers a unique opportunity to work directly
with our race-winning software stack and contribute to the continued advancement of autonomous racing technol-
ogy.
The goal of this thesis is to evaluate the planning concept of our autonomous racing stack by systematically
testing it within our existing deterministic testing framework. You will develop a scenario catalog to assess perfor-
mance across a range of static scenarios with non-interactive opponents, ensuring robustness in diverse race
conditions. Using established methodologies from autonomous driving evaluation — such as combined perfor-
mance metrics — the thesis will analyze how different parameter tunings influence the concepts performance and
safety. If time allows and the evaluation framework is fully operational, you may also explore the possibility of im-
plementing an alternative motion planning approach from existing research or developing your own, directly com-
paring it against the proven, race-winning concept used in our software stack. This work will provide valuable
insights into optimizing planning algorithms for high-speed autonomous racing and contribute to the future success
of TUM Autonomous Motorsport on the global stage.

Voraussetzungen

• Enthusiasm for autonomous racing
• Good programming skills in Python and C++
• Ability to collaborate in a team and engage in interdisciplinary research

Tags

FTM Studienarbeit, FTM Automatisiertes Fahren, FTM Buettner, FTM Informatik

Möglicher Beginn

sofort

Kontakt

Sascha Büttner, M.Sc.
sascha.buettnertum.de

Ausschreibung