Robust Object Detection for Off-Road Environments

Institut

Professur für autonome Fahrzeugsysteme

Typ

Bachelorarbeit / Semesterarbeit / Masterarbeit /

Inhalt

experimentell / theoretisch /  

Beschreibung

Object detection in off-road settings represents a core challenge for autonomous systems operating beyond structured, urban environments. Unlike well-mapped road scenes, off-road terrain introduces significant variability in visual and physical conditions, including unstructured vegetation, undulating or rough terrain, diverse lighting, seasonal variations. Traditional object detectors—primarily trained on urban datasets—often struggle in these conditions due to insufficient generalization, sensor noise, and a lack of contextually relevant training data.

There are several directions that you can choose from, depending on your interests and background:

1. Detector Development: Designing or adapting object detection architectures for unstructured, off-road environments.
2. Explainability: Making object detectors more interpretable and transparent for human operators and developers.
3. Online Monitoring: Develop monitoring modules that estimate confidence, uncertainty, or out-of-distribution (OOD) input detection.

 

Key Facts

Type:	MA/SA/BA, also for Informatics students
Starting Date:	Immediately
Supervisor:	Prof. Dr.-Ing. Johannes Betz
Advisor:	Yuchen Zhang, M.Sc
Programming Language:	Python
Language:	English

 

Work can begin immediately. If you are interested, simply send an email with your CV and academic transcript to yuchen2.zhangtum.de ;)

Tags

AVS Zhang

Möglicher Beginn

sofort

Kontakt

Yuchen Zhang
yuchen2.zhangtum.de