Recently, Sony’s AI research department announced the development of an autonomous table tennis robot named Ace. It completed high-level competitive testing in Tokyo, creating a new breakthrough in the fields of AI and robotics. It can compete head-on with top human table tennis players and even win multiple games. The related paper was published in the journal Nature on April 22. In terms of hardware configuration, Ace is equipped with nine synchronized cameras and three sets of vision systems, which can capture and track rotating table tennis balls at high speed, with high positioning accuracy and data processing speed.
At the same time, the R&D team created an eight-joint customized mechanical platform for it, which is also the minimum required configuration for competitive-level batting: three axes control the racket position, two axes adjust the racket angle, and the remaining three axes accurately control the hitting power and ball speed.
With this combination of software and hardware, the robot can calmly handle irregular ball paths such as rotating balls and scratching balls. According to the R&D team, as early as April last year in the competition, Ace's number of direct serve points was significantly ahead of the human elite players by 16:8.
However, as the battle report came out, the words "beating the top human players (Top-level)" appearing in Sony's promotion caused widespread controversy in the international academic community.
John Billingsley, a retired professor of mechatronics at the University of Southern Queensland in Australia, bluntly described Ace's winning strategy as "cracking walnuts with a sledgehammer" tactic.
He pointed out that human players only rely on their eyes to judge the path of the ball, while robots rely on nine high-speed cameras surrounding the table to achieve real-time analysis of 3D spatial data in the entire area without blind spots, relying on the huge information gap to crush their opponents.
In addition, Peters, a professor at the Technical University of Darmstadt in Germany, also commented that Ace can only complete the single closed-loop task of table tennis, and the technical scenario is highly limited.
Even if it performs well in table competitions, it cannot break through the technical bottleneck of general-purpose robots. There is still a huge technical gap in core problems such as fine grabbing and multi-scenario adaptive operation.
Related articles:
The table tennis robot Ace defeated top players in Tokyo and set the first place in the history of sports robots