Amazon company established its millionth robot in the order fulfillment center in Japan, which strengthened the company’s position as the world’s leading manufacturer and operator of mobile robotics. In the near future, there are plans to introduce a new artificial intelligence system that will make the entire fleet of robots even smarter and more efficient.
The new generative AI model was named DeepFleet. Her task is to coordinate the movement of robots through the entire network of Amazon logistics centers and reduce the average time of their movement by 10%. This will make it possible to deliver parcels to customers faster and reduce costs. The company explains that DeepFleet should be thought of as an intelligent traffic control system in a busy city that helps optimize routes and reduce delays — but in this case, it’s about navigating robots inside warehouses.
The technology was created on the basis of large volumes of data on the movement of goods at Amazon facilities, as well as using tools from the AWS cloud platform. Because DeepFleet is built around learning artificial intelligence, the system will become more efficient over time and discover new ways to optimize.
Amazon is also preparing to test the delivery of parcels by robots. For this, the company is completing the construction of a special “humanoid park”, where the tests will take place. The process itself will involve software based on artificial intelligence. For now, Amazon intends to use third-party equipment for testing.
