Inspiration
We have a passion for robotics and how they can be used to automate repetitive yet necessary aspects of business and human life. Both of us have worked retail and understand that taking inventory is a time consuming task. Taking inventory can easily become a monumental task. The larger the scale, the more time is consumed and more human resources are required.
What it does
ARI stands for Autonomous Recording of Inventory. It uses Augmented Reality tags to count and describe inventory. It can use the AR tags to determine how much inventory you have by traveling through the aisle of a warehouse. It automatically distinguishes between different inventory items and counts the number of stock (in our case, boxes).
How we built it
A computer uses a camera to locate the AR tags. Each AR tag gives a description of the product and our algorithm counts the occurrences of each tag. Furthermore the distance of each tag is calculated to tell whether or not there is more boxes hidden behind it.
Challenges we ran into
Our biggest challenge was making use of the resources we had at hand. Only one of our computers had a webcam which was required for live testing. Another issue was finding boxes to test this on. We made the best use out of the plenty of boxes around the arena and to help out our organizers.
Accomplishments that we're proud of
We're particularly proud of the two different modes our project exhibits live feed and scanning as well as being able to calculate the distance of an AR tag from just a webcam.
What we learned
Flask is an amazing tool to build powerful python applications quickly
What's next for Project ARI (Autonomous Recording of Inventory)
We would like to switch from AR tags to actual object detection and create a more organized system for more reliable counting to be scaled into a production setting.
Log in or sign up for Devpost to join the conversation.