This bachelor/master thesis will focus on interpreting sensor data, e.g. cameras and laser, from outdoor urban environments, to enable high level understanding of the scene. Specifically, the goal is to extract geometric and topological properties of different urban scenes as streets and crossings.

Such understanding would allow autonomous robots or cars to have a more sophisticated understanding of the environment, allowing them make more intelligent decisions.

 

Required skills:

Students should have a solid knowledge in C++ programming and ideally a good background or interest in statistical inference/machine learning.

 

 Contact:

Roderick de Nijs

rsdenijs@lsr.ei.tum.de