news

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

IT Home reported on August 11 that on August 7, Google's DeepMind announced that its research and development team has developed a table tennis robot that can reach the level of human amateur table tennis players in competitions.

▲DeepMind Project

According to reports, the results were evaluated through 29 matches between robots and humans.Robots won 45% (13/29)All players were human players that the robot had never seen before, and their skill levels varied from beginner to tournament level. While the robot lost all of its matches against the most advanced players, it won 100% of its matches against beginners and 55% of its matches against intermediate players, reaching amateur human level.

▲i-Sim2Real Project

IT Home noted that as early as 2022, Google announced its research on the "table tennis robot" project (i-Sim2Real), which can catch the ball 340 times in one round when playing against humans. Google claimed at the time that at the rate of AI growth, it would soon start against professional players.

Google decided at the time that the advantage of table tennis was that it was fairly tightly constrained (compared to playing basketball or cricket) and that it struck a balance between complexity and simplicity. In this process, machine learning models are taught what to do in a virtual environment or simulation, and then apply that knowledge, with the goal of playing catch rounds with humans for as long as possible without making mistakes.