
Table tennis robots are getting a lot better. They are becoming more effective at handling harder, faster shots. Take Ace for instance: it is an AI driven robotic system that can compete with professional players. It uses 12 high speed sensors, including three IMX636 event-based vision sensors. It tracks 40mm balls and measures spins over 9000 revolutions per minute.
Must see 🚁 racing & photography drone courses
How Ace Handles the Unexpected | Net Bounce Recovery | Sony AI
Ace relies on a deep Reinforcement Learning trained control policy. As the above video shows, it can also react well to the unexpected.
[HT]
*Our articles may contain aff links. As an Amazon Associate we earn from qualifying purchases. Please read our disclaimer on how we fund this site.












































