Google Deepmind unveils new AI models to control robots

Google Deepmind, the Google AI research laboratory on Wednesday Announced new AI models called Gemini Robotics Designed to allow real world machines to interact with objects, navigated environments, etc.

Deepmind has published a series of demonstration videos showing robots equipped with folding gemini robotics paper, putting a pair of glasses in a case, and other tasks in response to vocal controls. According to the laboratory, Gemini Robotics has been trained to generalize behavior through a range of different robotics equipment, and to connect the elements that robots can “see” with the actions they could take.

Deepmind says that in tests, Gemini Robotics has enabled robots to perform well -included in training data. The laboratory has published a slimming model, Gemini Robotics-Er, which researchers can use to train their own models for robotics control, as well as a reference called Asimov to assess the risks with robots powered by AI.