Aus Aifbportal
Wechseln zu:Navigation, Suche

Automated/Autonomous Driving - Machine Learning basierte Informationsfusion aus Kamera- und LiDar-Daten

Information on the Thesis

Type of Final Thesis: Bachelor, Master
Supervisor: Jens Weber
Research Group: Applied Technical Cognitive Systems

Archive Number: 4.765
Status of Thesis: Open
Date of start: 2021-04-12

Further Information

Data-driven algorithms are used for environment perception in automated driving. Within the project KI-DataTooling (, new methods for the (partially) automated annotation of training data will be researched. In order to achieve this, complementary information from camera and LiDar data will be evaluated with machine learning algorithms and subsequently fused together according with a "late-fusion" approach. This results in a labeling pipeline whose components include Deep Learning based object detectors (CNNs), clustering algorithms, and data association algorithms. In your work, you will investigate a part of this labeling pipeline and thus contribute to one of the most important research fields that will enable the broad use of Deep Learning.

TASKS You can expect:

  • Exploring the state of the art of research in the respective topic
  • Designing a prototypical system structure (interface definition, architecture, algorithm selection, ...)
  • Implementation of a prototype
  • Evaluation on known datasets and/or real data


  • An interdisciplinary research environment with partners from science and industry
  • A constructive collaboration with bright, motivated employees
  • A pleasant working atmosphere
  • Modern hardware
  • Openness to creative ideas


  • Theoretical and ideally first practical knowledge in the field of AI algorithms (CNNs, Clustering, data association)
  • Basic knowledge and programming experience C++ and Python as well as ROS and Linux
  • Sound English and German skills
  • High creativity and productivity


  • current transcript of records
  • CV


Jens Weber