Amazon has over 100 warehouses in the United States, with 45 centers and 50 delivery stations, all of which employ over 120,000 warehouse workers. However, even though the demand for labor is high in these large spaces, only a small fraction is taken up by robots.
And, according to The Verge, it’s going to stay that way for at least the next 10 years.
This is because robots are currently too “imprecise” and “clumsy” to perform certain tasks that require human dexterity. Robots are better for functions that require specific, repeatable programs that they can be programmed to complete.
But, if decision makers want robots to do something else, “expensive, time-consuming reprogramming is needed,” The Verge says. Or, decision makers might run into a case where the function they want is still only in the research phase, and not yet perfected and available for commercial use. “Even the simple process of identifying an object and picking it up without having been trained on that object before requires a series of complex, sophisticated software and hardware that does not yet exist in commercial fashion,” The Verge says.
But the robots are coming:
While progress is slow on developing robots to increase autonomy options for human workers, progress is still happening: “robots are starting to gain levels of vision and motor control that are approaching human-levels of sophistication,” The Verge says.
For example, labs are working on these advances, like the ones at UC Berkeley that are developing a low-cost robot equipped with humanoid arms controlled by a central system called Blue; the robot is able to “perform complex tasks,” such as folding laundry. A West Coast startup, Kindred, created a robotic arm that’s currently being used in Gap stores, which uses “a mix of human piloting and automation to perform dynamic product picking,” The Verge says.
Current Amazon warehouses use small robots called “drives” to deliver products to human workers by following a predetermined route created for that warehouse’s layout. But, even still, wider deployment of this solution, and others, still have years to go – hence the 10-year prediction by Scott Anderson, Amazon’s director of robotics fulfillment. “In the current form, the technology is very limited. The technology is very far from the fully automated workstation that we would need,” Anderson said in a previous interview with Reuters.