Image Credit: FLIR Systems
The AWM Frictionless™ shopping experience is a contactless, fully autonomous retail shopping system developed by AWM Smart Shelf. Using AWM Frictionless™, it is possible to enter a store, select an item and simply walk out of the store with the item of choice. The customer is charged automatically and receives a receipt for the item via e-mail.
AWM Frictionless™ employs FLIR Blackfly S GigE Machine Vision Cameras, enabling customers to be automatically charged for items as soon as they leave the store, removing the need to queue, scan, or physically pay for the items they choose.
The very first AWM Frictionless™ store was launch in Santa Ana, California, in March 2020 - a micro-market convenience store selling useful items such as sodas, water, snacks and refills on basic necessities including milk, soap, and hygiene products.
Toilet paper actually was the first purchase when we opened.
Kaitlyn Kempiak, Marketing Director, AWM
What Happens During an Autonomous Contactless Shopping Experience?
Before shopping, the customer downloads the store app and provides their payment information. They receive a QR code from the app, scanning this to be let through the store’s electronic doors.
Once they are inside the store, the customer simply shops as normal, and any products they choose are tracked and added to their digital basket. Once they have finished shopping, the customer can just leave the store.
Customers appreciate the convenience of having the store located inside their apartment building, especially during COVID-19 pandemic, there is no scanning, no speaking with a store associate, complete friction-free shopping.
Kaitlyn Kempiak, Marketing Director, AWM
The store aims to offer 24/7 opening eventually, enabling local residents and the general public to enjoy contactless shopping all the time.
Items in an AWM Frictionless™ store are tracked using a combination of FLIR Blackfly S GigE cameras and a deep learning algorithm that has been trained to accurately identify products before adding them to the customer’s basket.
Image Credit: FLIR Systems
The system will automatically remove an item from the virtual shopping basket if the customer returns this to the shelf. It is also able to keep track of a shopping bag, should the customer puts this down before moving to another aisle.
Why Did AWM Choose FLIR Blackfly S Cameras for Their Frictionless Checkouts?
When selecting the most appropriate camera system for Frictionless™, AWM opted for FLIR machine vision cameras. The company gave a number of reasons for this, including reliability, quality, helpful documentation, and first class customer support.
The Blackfly S cameras provide a number of key features that make them highly suited for AWM’s application; for example, the ability to adjust white balance to represent colors accurately and a color correction matrix to reproduce those same colors in a range of lighting conditions.
The system also offers the useful ability to process images from a large number of cameras quickly thanks to features like Packet Delay and Chunk Data Timestamp.
“To perform accurate tracking and 3D reconstruction across many cameras (32+ in some cases), it’s important we have accurate time information, down to the millisecond, for when frames were captured, and not necessarily the time when frames arrive at the computer for processing. This is especially true for GigE cameras operating over a network. The Chunk Data Timestamp feature on the Blackfly S cameras allows us to do this,” points out AWM.
Multiple cameras are installed on the ceiling, allowing them to effectively monitor the whole store. The system is also supported by weighted shelving, meaning that it is not possible to take advantage of the system by sneaking a small item, such as a pack of gum, into a sleeve or hand without this being seen by a camera.
The system has a record of precisely how much each item weighs, and this additional data input helps it to be more accurate and reliable.
Image Credit: FLIR Systems
The Color Correction Matrix is a crucial feature of FLIR Blackfly S cameras because it helps them to reliably reproduce colors in a wide range of lighting conditions. Ensuring that color is captured consistently is vitally important in Deep Learning applications because this increases the accuracy of the neural network and helps reduce the amount of training data required for the neural network to function correctly.
Touch-Free, Automated Retail Experiences in the Real World
The Frictionless™ experience is currently in place in micro-markets, and AWM has developed strong foundations to enable this technology to expand into other spaces, for example, convenience stores, traditional retail stores, and even supermarkets.
AWM Frictionless™ can be adapted to work in a range of existing stores, providing a contactless, fully autonomous shopping experience or adding a further, easy-to-use checkout option for customers.
The AI is continuously being trained to recognize and work with new products, and AWM is already working on a tool to recognize fresh produce ahead of deployment in grocery stores.
What has Been the Feedback from Customers?
Customers have reported that the Frictionless™ experience is incredibly easy to use while saving them time and reducing health risks. Kempiak also notes how a number of apartment residents have been making a visit to the micro-market a core part of their daily routine.
Image Credit: FLIR Systems
When the value of contactless checkouts, easier stock management, extended store hours, higher security, and reduced checkout times is considered and put into perspective, the AWM Frictionless™ experience is nothing short of revolutionary.
The machine learning solution continues to develop, making shopping easier to use and more convenient, and resulting in more stores and shopping formats becoming viable candidates for the Frictionless™ experience.
This information has been sourced, reviewed and adapted from materials provided by FLIR Systems.
For more information on this source, please visit FLIR Systems.