Analysis of Racing Interaction Using Time-Optimal Sampling Strategies for Motion Prediction in Autonomous Vehicles

Institut

Lehrstuhl für Fahrzeugtechnik

Typ

Semesterarbeit / Masterarbeit /

Inhalt

experimentell / theoretisch /  

Beschreibung

The TUM Autonomous Motorsport team develops software for the autonomous racecars of the Autonomous Challenge (IAC) and the Abu Dhabi Autonomous Racing League (A2RL). The team made history by winning the A2RL in 2024 at the Yas Marina Circuit against teams from international universities, competing at speeds exceeding 200 km/h. To achieve this, motion prediction, the prediction of the time-dependent vehicle positions within a time horizon, plays an important role in this software stack.

The aim of this thesis is to integrate an existing time-optimal sampling planner into the motion prediction module used in autonomous racing. This planner will generate optimal trajectories for opponents based on their predicted movements. The focus will be on analyzing how this integration affects racing interaction behavior, specifically in scenarios where both the ego vehicle and opponents continuously adjust their trajectories to avoid collisions.

Work packages:

• Literature review on the state of the art and research on time-optimal planning and motion prediction in autonomous racing
• Integrate the existing time-optimal sampling planner algorithm into the motion prediction module
• Analyze the interaction behavior between the ego vehicle and opponents using the integrated prediction module.
• Identify specific scenarios where the time-optimal planning approach does not work effectively and propose modifications or alternative strategies to address identified issue
• Discuss the findings from the interaction behavior analysis and scenario evaluation

 

Voraussetzungen

• independently familiarize yourself with the topic
• creativity
• a structured way of working
• knowledge of ROS2
• programming experience in C++ and python

Sprache

deutsch / englisch

Tags

FTM Studienarbeit, FTM AV, FTM Esser, FTM Informatik

Möglicher Beginn

sofort

Kontakt

Daniel Esser, M.Sc.
Raum: MW 3505
Tel.: +49 89 289 15342
daniel.essertum.de

Ausschreibung