news 
Nvidia is Helping Humanoid Robots Learn through Apple Vision Pro Instruction
At CES 2025, Nvidia CEO Jensen Huang unveiled a groundbreaking new development in the field of robotics. The company has partnered with Apple to enable humanoid robots to learn through Apple Vision Pro instruction.
This move is a significant step forward for the industry, as it allows humanoid robots to be taught new skills and behaviors using a simple and intuitive process. The technology, which Nvidia calls GR00T Blueprint, enables users to create digital twins of their actions by capturing them using an approximation of a real-life environment through Apple’s Vision Pro.
Once captured, these digital twins can then be used to simulate the robot’s actions in advance, allowing for more accurate and efficient training. The implications are far-reaching, and have the potential to revolutionize the field of robotics.
According to Nvidia, GR00T Blueprint is part of its broader effort to make humanoid robots a reality. The company has been at the forefront of this technology, announcing last year that it was developing a fleet manager for warehouse robots, called Mega.
The new partnership with Apple represents another major milestone in this space. Imitation learning is a fundamental aspect of teaching robots a new skill, and Nvidia’s collaboration with Apple ensures that this process becomes even more accessible and easy to use.
As the CEO of Nvidia highlighted during his CES keynote, imitation learning has a crucial role to play in educating systems designed to automate existing tasks. The ability for users to remotely teach robots through teleoperation is a game-changer, enabling instant digitization of actions.
It’s no surprise that Nvidia is leading the charge in this area, given its track record and industry leadership.
Source: techcrunch.com
