You have to see this tennis-playing robot to believe it
They did it with fragments of primitive human tennis skills
· TechRadarNews By Lance Ulanoff published 17 March 2026
Share this article 0 Join the conversation Follow us Add us as a preferred source on Google Newsletter
Get the TechRadar Newsletter
Sign up for breaking news, reviews, opinion, top tech deals, and more.
Contact me with news and offers from other Future brands Receive email from us on behalf of our trusted partners or sponsors
By submitting your information you agree to the Terms & Conditions and Privacy Policy and are aged 16 or over.
You are now subscribed
Your newsletter sign-up was successful
An account already exists for this email address, please log in. Subscribe to our newsletter
- Galbot found an unprecedented way, called LATENT, to train robots
- Using "skill fragments", they trained a Unitree G1 robot to play tennis
- The robot developed relatively robust tennis skills based on this minimal training
Future Wimbledons in which a fifth-seeded tennis pro squares off against a sixth-seeded robot just transitioned from the realm of science fiction to something that feels inevitable.
How did we get here? Blame Galbot and its LATENT innovation.
The best robot athletes, those that can do karate, box each other, or do parkour, are either remotely controlled or highly scripted to perform a pre-programmed set of actions. Real-time competition against, say, a human opponent is thought to be difficult or impossible. But now Galbot and a team of researchers have done it: used minimal learning to teach a Unitree robot how to play tennis against an unpredictable human opponent.
Article continues below
They call it "Learning Athletic Humanoid Tennis Skills from Imperfect Human Motion Data" (which they jury-rigged into LATENT). Instead of highly detailed robot training that captures the full range of human tennis skills, LATENT focuses on "motion fragments that capture the primitive skills used when playing tennis".
🎾🤖 INSANE! Researchers Can't Beat This #TennisRobot Anymore! #LATENT is Pro. #humanoid #robot #ai - YouTube
Somehow, the researchers figured out how to use these bits and pieces of tennis skills, or what they call "imperfect" data, to provide enough insight about "human primitive skills in tennis scenarios."
The robot, a Unitree G1, can then draw on those fragments to make sense of live gameplay and, according to researchers, "consistently strike incoming balls under a wide range of conditions and return them to target locations."
That's a dry way of describing what happens in the startling demonstration video in which a Unitree G1 robot adeptly plays — and sometimes outplays — a human tennis player.
Get daily insight, inspiration and deals in your inbox
Sign up for breaking news, reviews, opinion, top tech deals, and more.
Contact me with news and offers from other Future brandsReceive email from us on behalf of our trusted partners or sponsors