Google DeepMind Unveils AI Models Tailored for Robotic Applications
2025-03-13 / Read about 0 minute
Author:小编   

Google DeepMind has launched two advanced AI models, Gemini Robotics and Gemini Robotics-ER, aimed at empowering robots to tackle intricate real-world tasks. These models, which are built upon the foundation of Gemini 2.0, possess multimodal world understanding capabilities. This allows robots to grasp new situations and execute operations, such as origami folding and desk organization, without the need for tailored training. Specifically, Gemini Robotics-ER concentrates on refining the robot's spatial comprehension of the physical world and integrates with low-level control systems to introduce novel functionalities.