Google DeepMind Advances Robotics with New Gemini AI Model for Whole-Body Control
Google DeepMind has unveiled the latest version of its Gemini Robotics AI model, which now enables whole-body control of humanoid robots. This update allows robots to perform complex tasks requiring coordination from feet to fingertips, such as walking, crouching, and manipulating objects. The model also supports better dexterity with five-fingered hands, enabling tasks like sealing bags and unscrewing lightbulbs. Additionally, the Gemini Robotics ER model has been updated to improve task completion and safety, allowing multiple robots to work together and detect human presence more effectively. 
