Meet a Flying, Juggling Robot
The first time you see it, it’s almost too extraordinary to be believed. Best to watch it for yourself, first.
But while the video soon went deservedly viral—it’s garnered almost 2 million views on YouTube to date—few seem to have attempted to actually explain what’s at work here.
“It’s a proof of concept system,” Sergei Lupashin of ETH Zurich tells Technology Review, noting in an e-mail that the bots “definitely can’t play competitively against a human opponent”—yet. Each of the little quadrocopters carries the head of a badminton racket, he explains, and the ball tossed in the air is covered in retroreflective tape—like the kind safety-conscious cyclists use at night—to make it more easily spotted by the bots’ motion-capture system.
Each act of juggling follows a specific sequence: “the cameras see the ball/quadrocopters (at 200hz),” explains Lupashin, “send that information to an estimator/predictor, then there is some logic as to what’s currently happening (is someone intercepting the ball? who? where? when?) and this goes to a very high-performance controller that guides the quadrocopters to follow the appropriate trajectories.” The quadrocopters are equipped with inertial sensors that help it move with the dexterity and precision needed to achieve the feats seen in the video.
Lupashin further explains that while other roboticists working in similar fields are most interested in perfecting the sensors and localization capabilities of bots, the Zurich team is more interested in adaptability—teaching the quadrocopters to respond dynamically and successfully to that which it can’t predict or model.
Quadrocopters are being programmed by the Zurich group and others to do an astounding array of things—everything from constructing buildings to flying in formation to playing the piano.
Does Lupashin feel any pressure to compete with the University of Pennsylvania’s GRASP (General Robotics, Automation, Sensing & Perception) group, for instance—whose more serious-minded robots have, after all, chosen careers in construction rather than in juggling and musical performance? “I can’t say our system is better or vice versa,” he writes, “and that’s not the point! The point is we do things ourselves and then compare notes (we’re in quite close contact with some of the guys there) and everyone gains a lot of experience/ideas from having different approaches to compare.”
Keep Reading
Most Popular
Large language models can do jaw-dropping things. But nobody knows exactly why.
And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.
How scientists traced a mysterious covid case back to six toilets
When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.
The problem with plug-in hybrids? Their drivers.
Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.
Stay connected
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.