Robotik | Expert portrait
Dr. Thilo Rörig

Dr. Thilo Rörig wants to make the impossible possible. As a scientist, he knows the methodology that enables the precise construction of architectural models with complex, curved surfaces by using mathematics and geometry.


It is also important to know that architectural models are generally very precise. If, however, you want to build these complex designs, there will be slight deviations in the individual parts due to the manufacturing process. When assembled from many components, this deviation quickly adds up to a few millimeters. Algorithms now compensate for the lack of precision in implementation. At werk5 in Berlin, Thilo Rörig and Christophe Barlieb manage to adapt the precision of the implementation to the architectural design with a collaborative robot, Cobot for short. Virtual geometry is transformed into physical geometry -at the touch of a button. The mathematical foundations on which the calculations of the algorithms are based are only two years old. On the basis of the scientific research results, Thilo Rörig can generate components with a physical thickness on the computer and plan exactly how the robot will later assemble the parts with a corresponding thickness or curvature. The first proof of his work in real size is the Tricolumn, which is 17m high and stands on the campus of the University of Düsseldorf. Prof. Kruse's artwork was already judged as not doable by its creator with the words "From my experience this is not feasible. [...] When a composer writes something for a transverse flute that the flutist cannot play, it's going to be tough." Then came Thilo Rörig. Other architects have also had to bury their dreams and their designs have remained unrealizable visions of the future.
Thanks to modern mathematics and algorithms by Thilo Rörig and his colleagues, this should now be a thing of the past.


Thilo Rörig has a great passion for interdisciplinary cooperation between science and business. His motivation: Using mathematics in the real world. His motto in life: work scientifically and still solve real problems. Making things possible that do not seem possible. When asked how to get young people excited about mathematics, he says: "It takes a passion for everything abstract. You have to learn to combine abstract facts with real problems. You need the ability to imagine what you can do with abstract formulas. The fact that even as a scientist you sometimes have to poke in the proverbial fog in order to find solutions describes his approach to the concrete application of robotics for the construction of architectural models with minor deviations: If I aim for a loophole in a mountain, it is more efficient not to aim at the empty space but, for example, to the left of it, so that when I hit the mountain, I clearly know that the loophole is on the right. Together with Christophe Barlieb, he found his experimental field at werk5 in Berlin. Both have known each other for more than ten years. The cooperation at werk5 began with a ZIM-funded interface project between industry and research.


 


For the future, Thilo Rörig clearly focuses on man and machine: "Robotics will find its way into our everyday lives. Vacuum cleaner robots are just the beginning. There are tasks that humans like to hand over to machines and others that only humans can do. Robots and AIs will definitely change our working lives. But humans will not become superfluous. Why is this so? If you train a young person in three years, he is a specialist in his field and yet he can do so much more than a machine. Human versatility is not replaced by a robot. Only humans understand connections that are not directly given. This cannot be achieved by machine learning. In the end, Thilo Rörig, father of two children, has one more tip for the next generation: Do things for which you have a passion. If you do things with passion, you have found your place - whether as a carpenter, architect or scientist.


 


Further Links


Artikel auf tagesspiegel.de


Artikel auf wissenschaft.de


Projekt Tricolumn