-
-
Frontend
-
Arduino 1
-
Circuit diagram
-
Arduino 2
Website
https://getthegarbageouttahere.tech
incase camera does not open please use this link --- https://affectionate-ramanujan-20fb87.netlify.app
Inspiration
Are you tired of seeing trash everywhere? Do you want to see a change in the world? Well, this is the right spot to start! We wanted to build a robot that would be able to sort trash properly would allow for cleaner and more efficient environment. This would allow for less trash to be spread into oceans and around the planet. Our robot would roam around and using cameras and Machine Learning, which we created would be able to pick and choose waste and sort accordingly.
What it does
Given the resources and time, we have built the software and hardware components of the robot. This is also a virtual hackathon, so we didn't have the full resources and hardware to build a full robot. Our project allows for the sorting of random objects, into either Recycle bin, Garbage bin or Compost bin; similar to a local municipality in Canada. By utilizing Machine Learning through TensorFlow, we can easily detect if the object is categorized as trash.
How we built it
Using React.js as our Front-End and Node.js for the backend, we have developed a cutting edge solution to this environmental problem. Using the power of Arduino, we have implemented seamless communication between our web application and backend to turn on LEDs for corresponding trash (Recycle, Garbage, Compost). As mentioned above, Machine Learning through TensorFlow was used to detect objects using webcam and sort it accordingly.
How to test it?
Currently, we have trained few models for the demo, which includes, fork, spoon, orange, banana, apples, bottle, cup, chair, book, and a few more. Find the physical material for the best and accurate results, however, images can also be used.
What's next for Trash Sorter
With access to more hardware, we would like to create a robot which is able to actually move around and pickup objects. The robot can be based of the software and the hardware components, that was done so far.
Notes
This was done for the Environmental Challenge at YU Hacks.
Log in or sign up for Devpost to join the conversation.