Google says Gemini AI is making its robots smarter
What is Gemini AI used for in Google's robots?
Gemini AI is used in Google's robots to improve their navigation and task completion abilities. It allows users to interact with the robots using natural language instructions. The robots can watch a video tour of an environment, learn from it, and execute commands based on their observations. Gemini AI has shown a 90% success rate in following over 50 user instructions in a 9,000-square-foot area.
How does the Gemini 1.5 Pro enhance robot interaction?
The Gemini 1.5 Pro enhances robot interaction by enabling better navigation and task completion. Its long context window allows users to interact with robots using natural language instructions. Robots can learn about their environment by "watching" a video tour of a designated area and then undertake commands based on their observations using verbal and/or image outputs. This has led to a 90% success rate across over 50 user instructions in a 9,000-square-foot operating area.
What method is used to teach robots about environments?
The method used to teach robots about environments is through the use of Gemini AI. Researchers film a video tour of a designated area and use Gemini 1.5 Pro to make the robot "watch" the video to learn about the environment. This allows the robot to undertake commands based on what it has observed using verbal and/or image outputs.