Can Robots Replace Cashiers

The effort to drive more automation in the retail industry has been on for more than a decade. There have been some successes so far, such…

The effort to drive more automation in the retail industry has been on for more than a decade. There have been some successes so far, such as, the automatic checkout counters we see now a days in grocery stores. More recently, such counters made headway into fast food joints, such as McDonalds. Then Amazon Go came by, and so did other competitors who wanted to eliminate the experience of waiting for a store clerk to process payments. It is predicted that 97% cashiers will loose jobs to some form of automation in the next 2 decades. Here is a summary of the recent state of art in retail automation.

However, a key inhibitor to this automation still exists. How will the Mom and Pop’s, small restaurants, and local joints, who cannot afford expensive technology used in Amazon Go? They still need to be competitive with The Amazon’s, else enter the slow death spiral. As an example, virtually every store had to adopt credit card processing because of the onslaught of the cashless society. Similarly, the advent of people-less checkouts is destined for sure.

We are about to enter a world with ubiquitous robots; and it is my belief that robots will play a big role in automated checkouts. Robots have many subtle advantages over the current generation of automated checkouts. First, robots can be made to be much more personal and tailor to your emotions. Next, robots can be designed to interact with human beings using multiple senses: sight (Compter Vision), audio (Speech Recognition), touch (Capacitive Sensors). Such technologies enable robots to reproduce and replicate complex human actions in a far more profound way than computers ever could. Supplementing all this is deep learning fed by big data which make these technologies extremely potent.

So here is a prototype to provoke your thoughts. We use Anki Vector as the cashier (Anki made a host of popular robots and toys before shutting down earlier this year). You would notice that the cashier detects a human approaching it, scans items by reading custom images pasted on them (a better version can read bar codes), generates a QR code to process a mobile transaction. The user can scan the QR code with their mobile app, and be done… Potential solutions can include face recognition, and prompting the user for items they might have forgotten based on past shopping lists, the potential is huge.

Thoughts, comments… Please let us know. If you want to check the code, here is the git. And a clap of your hands will really help us. Follow the publication to know more about our efforts in this space.

If you want to learn more about Vector, or learn AI with Vector, I have a new course at: https://robotics.thinkific.com