Virtual Clean up
Trash Detection
Object Detection

WALL-E

Inspiration

Our inspiration was to help make the world a cleaner place, whether that be physically or virtually. The name "WALL-E" was inspired from the iconic character who was dedicated to cleaning up Earth after it was overwhelmed by waste. Just like that WALL-E, this project embodies the spirit of restoration and care for the environment, but it extends beyond just the physical world. Our WALL-E is designed to clean both real-world waste and digital clutter, reflecting the idea that messes—whether physical or virtual—need to be addressed with the same level of attention and responsibility.

What we learned

Developing WALL-E brought many lessons. We improved our understanding of arduinos and hardware as well as our skills in motion tracking with the use of OpenCV and machine learning models.

How we built it

We started WALL-E with our backend. We made use of ML models in order to generate our categorical schemas which are based on the content of the user selected directory. For the physical WALL-E, we made use of OpenCV and a machine learning model in order to detect the waste items and using that input to communicate with the arduino.

Challenges

One of the main challenges we faced was getting the trash detection working correctly. For the backend, we faced the challenge of the application deleting empty folders of a directory as well as the files that are deemed as old or useless. These challenges led to constant trial and error and modifications to implementation, but were all worth it by the end.