Google DeepMind Unveils Gemini Robotics AI Model for Whole-Body Robot Control

Google DeepMind has announced an upgrade to its Gemini Robotics AI model, allowing humanoid robots to perform a wider range of actions. The new version, Gemini Robotics 2, builds upon the previous model's ability to control a humanoid robot's upper body by supporting "whole-body motions" from the feet to the fingertips. This advancement enables robots to engage in various physical activities, including walking, crouching, stretching, and manipulating objects.

According to Google DeepMind, the Gemini Robotics 2 model has been demonstrated in collaboration with Apptronik, a robotics company, using their Apollo 2 robot. Videos showcasing the robot's capabilities in action have been shared by Google, highlighting its ability to perform complex tasks such as bending over to pick up a watering can and retrieving specific items from a shelf. The Apollo 2 robot is seen taking a baseball glove off a shelf, exemplifying the robot's newfound capacity for whole-body motions.

The Gemini Robotics 2 model's capabilities have the potential to revolutionize the field of robotics, enabling humanoid robots to assist humans in various tasks and applications. As researchers continue to develop and refine this technology, we can expect to see more sophisticated and versatile robots that can interact with their environment in a more human-like manner.