Google DeepMind has introduced Gemini Robotics 2, a significant update that enhances its robotics capabilities. This new version expands beyond previous iterations by incorporating full-body movement control for humanoids, advanced manipulation for multi-fingered hands, and improved collaboration among multiple robots.
The importance of Gemini Robotics 2 lies in its ability to unify locomotion and manipulation, addressing a common limitation in robotics. As Carolina Parada, Head of Robotics at Google DeepMind, stated, the new model enables robots to think, act, and interact intelligently, which is crucial for tackling complex tasks in diverse environments.
Looking ahead, Gemini Robotics 2 is now available through Google AI Studio and in private preview on the Gemini Enterprise Agent Platform. The rollout of the core VLA and On-Device 2 models to early-access hardware partners is also underway. No further timeline was disclosed at the time of publication.
Editor's Note
The launch of Gemini Robotics 2 by Google DeepMind represents a significant advancement in the integration of AI with robotics. By focusing on whole-body dynamics and multi-robot collaboration, this development could enhance operational efficiency in various sectors, including logistics and manufacturing. The emphasis on complex end-effectors also suggests a shift towards more versatile robotic applications in unstructured environments.
Leave a comment