Sony AI developed an autonomous robot system capable of competing with elite human table tennis players in real-world matches.
Research published this week in the journal Nature details Ace, a robotic system combining high-speed sensing, reinforcement learning—a machine learning training method based on rewarding desired behaviors—and precision hardware.
The system tracks the ball, anticipates opponent shots, and executes returns under real competition conditions. Ace operates fully autonomously without human intervention during play. The robot won three out of five matches against elite amateur players.
The system struggles against professional-level players, winning only one out of seven games in those specific match-ups. Table tennis demands split-second visual perception, millimeter-level motor control, and rapid strategic adaptation. The system utilizes nine active pixel sensor cameras to determine the ball’s precise three-dimensional position.
This optical setup allows the robot to estimate the ball’s spin and axis of rotation within milliseconds. Researchers note this technology establishes capabilities necessary for robots operating in dynamic physical environments.
Cover Photo by Imad Clicks