@ctorobotics: This robot learned to play air hockey. 🏒🤖 It was trained entirely in simulation using reinforcement learning then transferred directly to a real robotic air hockey table. The system can react in around 20 milliseconds and achieve roughly 1 mm path-following precision. No human demonstrations. No hard-coded gameplay. Would you play against it? 👀 🎥 Media: Hudson Nock, Ian Hartley & Mauro Ferraz — The University of British Columbia ( UBC ) ⚠️ This content is shared for informational purposes only. CTO Robotics Media is a media platform and does not own or develop the technology shown. Credit belongs to the original creators. #Robotics #AI #ReinforcementLearning #RobotLearning #ComputerVision #Engineering #PhysicalAI #FutureTech