AI table tennis robot learned to play in just 90 minutes

A table tennis-playing robot can play a fair part in a rally against humans – but like many amateur players, struggles when attempting fancier shots. Yapeng Gao, Jonas Tebbe and Andreas Zell at the Eberhard Karls University Tübingen, Germany trained a machine learning algorithm to learn how to play table tennis. That algorithm then sent signals to a robot that allowed it to accurately play shots to within an average of 24.9 centimetres of the intended location. Read more: