You are using Internet Explorer and will not be able to use our website properly. Please change to an up-to-date browser for ideal presentation of the website.
"In robotics, I can let off steam mathematically".
Equipping young talents for the future, getting them excited about STEM subjects and introducing them to new technologies is a mammoth project. At KUKA, we are working on this in various places and are in close contact with educational institutions and politicians to enable the younger generation to get to grips with new technologies.
25 November 2022
Reading Time: 4 min.
KUKA's commitment this November also includes hosting the awards ceremony for the national mathematics competition. The young prize winners from all over Bavaria met at KUKA headquarters in Augsburg. When they receive their prizes at the end of November, the boys and girls will have completed 14 months of algebra, geometry and combinatorics. The National Mathematics Competition is a school competition for everyone who is interested in mathematics. It consists of two homework rounds and a mathematical discussion in the final third round. In addition to mathematical school knowledge, motivation and perseverance also count. One person who works at KUKA and has a PhD in mathematics is Jonas Schwinn. We talked to him about what he likes about the mixture of numbers and technology and what he would give young talents to take with them.
Jonas, at KUKA you have already been working for several years in the research area on topics such as machine learning or data analysis. How do your PhD in mathematics and your day-to-day work fit together?
Jonas Schwinn: "Robotics is a wonderful field of application for applied and theoretical mathematics. It's nice to be able to do interesting mathematics one moment and see the results in the form of a moving robot the next, which - thanks to mathematical considerations - is already doing its job a little better.
To move a robot, you have to know which axis angles have to be set so that the TCP - the front tip of the robot - moves to a desired position. This works by solving linear equations, and that is, after all, something that is partially learned at a young age in schools."
What was your focus for your math PhD?
"My focus in my studies was mathematical optimization. This is a wonderful field to work in robotics, especially in robotics research. Mathematical optimization methods are an important part of modern control algorithms. They determine how a robot should optimally move in order to solve a certain task. Here's a brief example: The robot is supposed to follow a line at the maximum possible speed while transporting a plate and ensuring that no food falls off. And formulating precisely such real-world problems in mathematical models: That characterizes my everyday work at KUKA. And my completed doctorate helps me immensely in this. I analyze data from the field of machine learning - a sub-form of artificial intelligence - and put it into mathematical models to find solutions to ensure that no food falls off the plate when a robot is transporting it."
Are there any other examples?
"Definitely. For example, we have developed an algorithm that decides how a robot should take an object out of a box in order to lose an object as rarely as possible during the gripping process. That sounds a bit trivial now - for a human it is, because eyes and mind interact ideally, and we do it unconsciously. For a robot, however, this is a major task.
To understand how machines learn, mathematical understanding is essential. Other applications from applied stochastics or statistics can be found in environment modeling. For example, when I have seen an object from different perspectives with a camera and now want to know which is the most probable position of the object.
In mobile robotics - that is, when our robots are mounted on mobile platforms to move freely in space - statistics come into play to create maps of the environment or to find out the exact position of the robot.
Theoretical mathematics also plays a big role in solving problems in robotics.”
You can definitely hear a great passion there: What fascinates you about numbers and technology?
"In general, I am a problem solver. Statistics and mathematical modeling and optimization help me to understand problems and develop solutions. I'm also fascinated by developing things that take over tasks from humans. That can be software or hardware - in robotics you do both."
What would you like to pass on to young talents?
"One of the most pressing questions is how we can make our society, and especially our industry, more sustainable and ecological in the coming decades. As mathematicians, you can play a central role in this, in that you and we together can plan the use of energy mathematically in the best possible way. This is a great ability - and somehow also a piece of our responsibility. The important thing is to keep trying out how you can turn mathematical knowledge and talent into solutions for real problems. And: never lose your passion - it's a great subject area."