IT之家 AISource publication

Hitch Open physical intelligence open-source platform was released at the CII Expo. A 142-bit refresh of the AI robot table tennis match…

IT Home On November 8, according to Lingang Group, on November 7, the 9th China International Import Expo featured the world’s first showcase in the field of physical intelligence.

IT Home On November 8, according to Lingang Group, on November 7, the 9th China International Import Expo featured the world’s first showcase in the field of physical intelligence.

Hitch Open, a global physical intelligence open-source platform from Silicon Valley in California, USA, and with its academic foundation from the University of California, Berkeley, held a global launch event in the exhibition area of the Port of Shanghai Expo.It officially released the physical intelligence platform to the world as an open source product.

According to IT Home, a乒乓球 match challenge involving an AI autonomous decision-making robot was completed on site. Created jointly by the team of Professor Gao Yue from Shanghai Jiao Tong University — Shanghai Chuangzhi College and the Hitch Open platform, it achieved a new 142 points record, breaking the world record for this project.

According to the introduction, Hitch Open is a physical intelligence competition platform designed for real physical environments. It aims to address the core challenges of artificial intelligence in complex physical worlds—such as perception, decision-making, and action—through extreme events in real-world settings, and to promote the development of autonomous driving, embodied intelligence, and robotics technologies.

And table tennis matches are a “limit test ground” for physical intelligence. The table tennis ball moves fast and has complex spin changes, leaving only a hundred milliseconds for the robot’s reaction window. This requires the robot to complete real-time analysis within the “perception — decision — action” closed loop. Each additional round represents a simultaneous leap in the robot’s perception accuracy, decision speed, and motion control capabilities.

Original source

IT之家 AI

Content notes

Original publication and rights belong to the source.

Machine translation · Refer to the original