What Is Amazon Go?


The robots haven’t taken over…yet. Amazon has unveiled its plans on its new project, Amazon Go. Amazon Go is a real grocery store that has no checkout lines of any kind. You essentially walk around in the store, pick what you need off the shelves, and walk right out. Just like that.

There are no cashiers, no lines, no cash or credit card exchanges, and no waiting. Right now there is only one Amazon Go location in America and it is in Seattle, Washington. It is basically a store where they are testing out all the kinks, as it is only 1800 square feet and holds simple items such as bread, milk, and cheese. The store is only available to Amazon employees in beta, with a public launch set for early 2017.

Customers will scan a 2d barcode on their phones as they enter the store to start the process. Anything taken from the shelves will be automatically added to the customer’s virtual cart, as their Amazon account will be charged when they walk out of the store. This technology combines machine learning, artificial intelligence, computer vision, and fusion technologies, very similar to driverless cars.

RFID technology will also be used, using radio waves to identify and track objects in a form of wireless communication. In a RFID system, a tag is attached to an item to be tracked and is connected to an antenna. This tag contains memory which stores the product’s electronic product code.

This RFID system is essential to this entire project. This system is essentially a system for tracking removal or placement of items at inventory location with a materials handling facility. So there will be cameras mounted on the ceiling, the walls, and shelves. There will be depth sensing cameras involved, as well as multiple sensors such as pressure sensors, infrared systems, scales, volume displacement sensors, and light curtains. Here’s a quick breakdown:

  • There will be image sensing devices on the shelves that will sense hands as they pass through a vertical plane set on the shelf. The management system can tell whether is it a behavior to remove or place an item.
  • The item is placed at a particular location on the shelf, with sensors mounted at a particular shelf location can tell which product is picked up.
  • The human recognition is picked up by the customer’s location provided by the their cell phone. If a customer is standing within a small area and picks or places back an item, they are associated with the product. An infrared sensor can be used to distinguish between a customer’s hand and inventory items.

This technology is obviously super revolutionary and has major consequences. If it picks up within the next decade, the jobs in the supermarket field will plummet. We are entering a new phase of having to balance technological advances with real, economic consequences.