Forgot your password?
Close
typodupeerror
Robotics

Ping-Pong Robot Makes History By Beating Top-Level Human Players (reuters.com) 29

Posted by BeauHD from the state-of-the-art dept.
Sony AI's autonomous table-tennis robot Ace has become the first robot to compete against top-level human players. Reuters reports: Ace, created by the Japanese company Sony's AI research division, is the first robot to attain expert-level performance in a competitive physical sport, one that requires rapid decisions and precision execution, the project's leader said. Ace did so by employing high-speed perception, AI-based control and a state-of-the-art robotic system. There have been various ping-pong-playing robots since 1983, but until now they were unable to rival highly skilled human competitors. Ace changed that with its performances against human elite-level and professional players in matches following the rules of the International Table Tennis Federation, the sport's governing body, and officiated by licensed umpires.

The project's goal was not only to compete at table tennis but to develop insights into how robots can perceive, plan and act with human-like speed and precision in dynamic environments. In matches detailed in the study, Ace in April 2025 won three out of five versus elite players and lost two matches against professional players, the top skill level in the sport. Sony AI said that since then Ace beat professional players in December 2025 and last month. "The success of Ace, with its perception system and learning-based control algorithm, suggests that similar techniques could be applied to other areas requiring fast, real-time control and human interaction -- such as manufacturing and service robotics, as well as applications across sports, entertainment and safety-critical physical domains," said Peter Durr, director of Sony AI Zurich and leader for Sony AI's project Ace.

The findings have been published in the journal Nature.

Ping-Pong Robot Makes History By Beating Top-Level Human Players More | Reply

Ping-Pong Robot Makes History By Beating Top-Level Human Players

Comments Filter:
  • Ok, but can it beat me at Wii Sports Table Tennis?

  • Run Forrest... (Score:4, Funny)

    by bosef1 ( 208943 ) on Wednesday April 22, 2026 @06:22PM (#66107688)

    So between this robot, and the robot that recently set a record half-marathon time, does this mean we can now build an android Forrest Gump?

  • Not that great an achievement (Score:1)

    by Anonymous Coward

    I just invented an AI that wins every point. On AI's serve, the ball is shot out of a gun at a speed that is impossible to return. When it's human's turn, AI switches on a gigantic fan so that human can't get the ball over the net.

    • Maybe you should first invent an AI that teaches you the rules of ping pong?

    • Re: (Score:2)

      by dvice ( 6309704 )

      Using guns and fans is forbidden by the law of table tennis. You would have to make a robot that uses paddle to hit the ball. When doing that, you have physics against you, so at most you might get something like 100 m/s speed for the ball. Faster than that and ball gets crused. With that speed it would be hard to return, but not impossible (it is about 3 times the speed humans can produce). The ball loses speed quickly, so standing behind the table might do the trick.

  • Does the robot rely on vision only with cameras in same position as eyes? Or is it using more sensors etc than the human

    • Re:Any videos? (Score:4, Informative)

      by vux984 ( 928602 ) on Wednesday April 22, 2026 @08:01PM (#66107816)

      If only there was some way to get at this sort of information, some article containing it.

      Ace's architecture integrates nine synchronized cameras and three vision systems to track a spinning ball with exceptional accuracy and speedy processing time.
      "This is fast enough to capture motion that would be a blur to the human eye," Dürr said.

    • Exactly. A humanoid form robot with restriction of no more than two cameras spaced no further apart than human eyes, that would be impressive. But without such restrictions, this is totally doable in Python, no need for AI. There is a guy who did some years ago a similar achievement of darts you can not miss the center, I bet he could do this too, given enough resources, even when this requires a bit more, like angles and collision moment speed and direction.
      • While the humanoid form is a needed restriction, I don't think restriction on sensors make sense (as long they are embedded). The end goal is to have robots fitting our human made world so they can replace worker and kill civilians not fair competitors at random sports. Even the joints of humanoid robots are way superior than ours in many area.

      • Re: (Score:1)

        by gtall ( 79522 )

        So Python can produce servo-motors? Amazing!!

      • Re: (Score:2)

        by allo ( 1728082 )

        But the cameras need to be able to move very quickly, change focus quickly, and have high resolutions images processed quickly. I think giving the robot the advantage of a few more cameras when it cannot have cameras evolved over millions of years is a good compromise.

    • Re: (Score:2)

      by ffkom ( 3519199 )
      A video might give away that their definition of "Top-Level" players just meant slightly-better-than-agerage players. Also the results are less impressive when you consider all the special infrastructure required to be installed for the robot to use. It wasn't like just a normal table tennis table somewhere and a two-legged autonomous robot.

  • Or maybe nobody has tried very hard to play ping pong until now. The Ukrainians can shoot down hundreds of drones and missiles per day. Those things move at hundreds of miles per hour and are programmed to actively avoid interception. I'm sure if the Ukrainians didn't have more pressing matters at hand they could deflect a single ping pong ball that follows Newtonian laws.

  • yes, they can be designed do specific tasks better than people - ever hear of the industrial revolution??

  • Playing against an adversary involves studying its weaknesses. We humans are just too slow. The next phase would be other robotics company building a contender. Can't wait.

  • The trend is robots becoming better in the physical environment. Next, you'll see a humanoid robot master ping-pong. The next step is put top-line LLM models in robots and they'll approach general-purpose capability. They will be like the first crank--start cars that couldn't go much faster than a horse. But eventually they replaced all those horses. Guess who is the horse now?
  • If conservatives have moral panic scapegoating problems with all ~5 trans athletes competing in women's sports like it's the apocalypse, where's the consistency of outrage for totally outclassing robots competing unfairly in human sports? Put robots in their own leagues or classes just like most every sport.

Slashdot Top Deals

A boss with no humor is like a job that's no fun.

Close