Safe and Fast Tracking on a Robot Manipulator: Robust MPC and Neural Network Control

Description

I worked on this project in the scope of my Master’s thesis for my degree at ETH Zurich. Our goal was to extend previous work on robust MPC and approximate MPC, and to demonstrate it on a real world robotic system in order to underline its practical significance. This work was conducted at the Max Planck Institute for Intelligent System under the supervision of Sebastian Trimpe. Furthermore, I received additional supervision from Johannes Koehler, who is a doctoral student at the University of Stuttgart. 

Our results were published in the robotics and automation letters (RA-L) and were further presented at the (virtual) International Conference on Robotics and Automation (ICRA).

Videos