Google’s Gemini AI model has proven its capabilities once again, this time with the help of a robot from Google’s Everybody Robots Division. Despite the division being shut down last year, the robots still remain, and Google decided to showcase the potential of Gemini by teaching one of these robots how to respond to commands and navigate the DeepMind office space.
To achieve this, Google utilized vision language models (VLMs) that are trained on images, videos, and text. These VLMs enable the robot to answer questions and perform tasks that require perception.
In a video demonstration, a Google employee asks the robot to take them to a place where they can draw things. The robot takes a moment to think and then guides the employee to a whiteboard. In another video, the robot is instructed to follow the directions on the whiteboard, which displays a map leading to the Blue Area. Impressively, the robot successfully follows the directions and proudly announces, “I’ve successfully followed the directions on the whiteboard.”
This demonstration highlights the advanced capabilities of Google’s Gemini AI model and its potential applications in various industries. With the combination of VLMs and robotics, Google has opened up new possibilities for robots to navigate and interact with their environment based on visual and textual cues.
The use of VLMs provides the robot with a deeper understanding of its surroundings, enabling it to process visual information and respond accordingly. This not only improves the robot’s ability to comprehend and interpret commands but also allows it to navigate complex spaces with ease.
The integration of AI models like Gemini into robotics showcases how technology continues to advance and enhance our daily lives. While this particular demonstration may seem small in scale, it signifies a significant step forward in the realm of human-robot interaction.
As AI models like Gemini become more sophisticated and capable, we can expect robots to play increasingly pivotal roles in various industries, such as healthcare, manufacturing, and logistics. These robots can assist with tasks that require perception, navigation, and problem-solving abilities.
In conclusion, Google’s recent demonstration with the robot from its Everybody Robots Division showcases the power of their Gemini AI model. By combining VLMs with robotics, Google has enabled the robot to respond to commands and navigate its environment successfully. This development paves the way for future advancements in human-robot interaction and opens up exciting possibilities for the integration of AI models in various industries.