The robot that can outplay elite table tennis players | Science, Climate & Tech News

A robot has become so good at playing table tennis that at times it defeated elite human players, in what has been hailed as “a longstanding milestone for AI”.

The robotic arm, built by Japanese electronics giant Sony, is a paddle-wielding robot by the name of Ace.

Ace was pitted against professional athletes and was found to have given them a real challenge with its nine camera eyes positioned around the court.

The robot, which also has eight joints that direct its movements, learned how to play the sport using the AI method known as reinforcement learning.

“There’s no way to program a robot by hand to play table tennis. You have to learn how to play from experience,” said Sony AI researcher Peter Dürr, co-author of the study published in the science journal Nature.

Sony said it is the “first time a robot has achieved human, expert-level play in a commonly played competitive sport in the physical world – a longstanding milestone for AI and robotics research”.

Michael Spranger, president of Sony AI, said: “Speed is really one of the fundamental issues in robotics today, especially in scenarios or environments that are not fixed.

“We see a lot of robots that are in factories that are very, very fast.

“But they’re doing the same trajectory over and over again. With this technology, we show that it’s actually possible to train robots to be very adaptive and competitive and fast in uncertain environments that constantly change.”

Read more from Sky News:
Pet owners warned of new EU travel rules
This is why your hay fever is getting worse

An Olympic-size table tennis court was built at Sony’s headquarters in Tokyo to conduct the experiments, so professional and other highly skilled athletes had a “level playing field” with the robot, Mr Dürr said.

Mr Spranger said it was important the robot didn’t have too unfair of an advantage – its speed, arm’s reach and performance were made comparable with a skilled athlete who trains at least 20 hours a week.

It plays by official table tennis rules on a typically sized court.

“It’s very easy to build a superhuman table tennis robot,” Mr Spranger said.

“You build a machine that sucks in the ball and shoots it out much faster than a human can return it.

“But that’s not the goal here. The goal is to have some level of comparability, some level of fairness to the human, and win really at the level of AI and the level of decision-making and tactics and, to some extent, skill.”

That means, he said, that “the robot cannot just win by hitting the ball faster than any human ever could, but it has to win by actually playing the game”.