Mapping and Localization for an Autonomous Racecar

In my first big software project during my studies, I developed the Localization and Mapping algorithm for the autonomous racecar “gotthard driverless”. The algorithm fuses estimates from camera, LiDAR and inertial sensors in a late-fusion approach into a coherent map.
We were able to publish a paper about our work at the International Conference of Robotics and Automation (ICRA19), titled Redundant Perception and State Estimation for Reliable Autonomous Racing.

Formula Student was my big passion. It is an international student competition, where teams from all around the globe build race cars and compete for the big prices in Germany, Austria, Spain or the UK. 

This video introduces the paper we published for ICRA19. In the beginning it shows some live action of the competition in Germany.