I am currently a guest scientist at MPI for Intelligent Systems, where I conduct research on safe robot learning. Previously (2014-2018) I was a PhD student working with Jan Peters in the Robot Learning Lab, where two 7-DoF Barrett arms are facing each other, awaiting eagerly for the day when they can beat each other in a deadly robot table tennis match. So far that has not happened! I think that in order to learn from scratch, one needs to look at how robots could learn safely without extensive prior knowledge or assumptions!
During my PhD I was working on optimal trajectory generation for table tennis and learning control algorithms (that are applicable to table tennis). I am especially interested in optimal control and reinforcement learning.
Learning control Table tennis Robotics Optimal Control Optimization
Our goal is to understand the principles of Perception, Action and Learning in autonomous systems that successfully interact with complex environments and to use this understanding to design future systems