The future of Amazon's logistics network will undoubtedly include artificial intelligence and robotics. However, it is always open that the AI drive machine will do the majority of the work. Scott Anderson, the point at which the Amazon warehouse is complete and an automated end-to-end automation system is implemented, takes at least 10 years. Anderson's remarks today released by Reuters emphasize the current speed of automation in environments that are familiar with robot labor, such as the Amazon warehouse.
Robots in today's labor force are proficient in repetitive and specific tasks that are accurately programmed. Robots need expensive and time-consuming reprogramming to get them to do other things. Robots that operate in a dynamic environment where robots can perform many different tasks and in which the robot must see and understand the environment is still in the area of research and experimentation. Even the simple process of identifying an object and selecting an object that has never seen it before requires a complex set of sophisticated software and hardware that is not yet commercially available.
While robots can help build the body of microchips and Tesla cars, human tasks such as warehouse operations are needed to be performed. At Amazon's facilities and service centers in other companies, massive labor is still carried out by human hands because it is difficult to see and train robots around the world and to use gripper grippers with human dexterity.
However, as part of an ongoing learning revolution that has accelerated the advancement of artificial intelligence research over the past decade, robots have begun to gain visibility and level of exercise control that approach human sophistication. Amazon is one of those companies that has pioneered such robots, and after the warehouse period picks up an object and moves to another part of the logistics chain to promote progress on the spot, so-called picking is done annually.
Many other companies and institutes are also making progress in this area. At UC Berkeley, there is a robot laboratory that has made significant progress in the field, and the new low-cost robot, a humanoid arm robot controlled by a central system called Blue, can perform complex manual tasks to fold towels thanks to AI – a vision system. Lab OpenAI used AI training technology, known as reinforcement learning, to teach robotic hands more precise and elegant movements, the types of movements that robots need to replicate humans in a warehouse. A startup based in San Francisco, Kindred creates a robotic arm called Kindred Sort and places it in a warehouse for retail gaps to combine maneuvering and automation to perform dynamic product selection.