Strolling through a Robot Half Marathon

It was a typical Friday afternoon when team robot control decided they needed a breath of fresh air. So they hopped onto a plane to Beijing for a week and came back as winners of the best international team award at the Beijing E-Town Half Marathon on April 19, 2026.

Yes. A robot half marathon. 21 kilometers, 124 teams from across the globe, and one small team from the Technical University of Munich who somehow pulled off the impossible.

The odds were truly stacked against us. Two weeks of preparation time, and we wouldn’t be getting our hands on the actual robot we’d race — the Tienkung Ultra — until we landed in Beijing. That meant our entire preparation had to happen in simulation. Luckily, simulating humanoids is a core competence of our team. But simulating and training a robot we didn’t even have in the lab, to run a long-distance marathon, in an unknown environment? That was a new kind of challenge.

Once in Beijing, we had just three days of actual hands-on time with the real robot. Three days to bridge the sim-to-real gap, tune everything for a 21-kilometer course, and make sure the robot didn’t overheat along the way. It certainly wasn’t a walk in the park.

Then came race day. 39th out of 124 teams.

Half the field dropped out, but we crossed the finish line. And not only did we finish — we did it as the best international team. No water cooling of motors, only efficient control to minimize motor load and keep the robot going until the very end.

Shoutout to the team on the ground in Beijing: Prof. Dr. Gordon Cheng, Dr. Julio Rogelio Guadarrama Olivera, Simon Armleder, and Xiangyu Fu!

What does this mean for STROLL?

We may or may not have attached some sensors to the robot for a little stress testing over the course of whole marathon… 😉

Whatever a typical Friday afternoon at team robot control looks like from here on out, we have a feeling Beijing won’t be the last stop.

Link to the original TUM article: https://www.tum.de/en/news-and-events/all-news/press-releases/details/tum-roboter-gewinnt-best-international-team-award