SmartSort: AI-Powered Recycling Bot SmartSort is an AI-driven recycling assistant that automatically detects recyclable items like plastic bottles and aluminum cans, helping promote smarter and cleaner recycling.
📚 Inspiration Walking around campus, we noticed many recyclable items being tossed into the trash. We wanted to create a system that uses AI to recognize recyclable materials and sort them correctly, making recycling easier for everyone.
🚀 What It Does Uses a camera and a trained YOLOv8 model to detect objects.
Classifies items as:
Plastic Bottle
Aluminum Can
General Waste
Activates a servo mechanism to direct items into the recyclable bin.
Lights up an LED indicator when a recyclable item is detected.
🛠️ How We Built It Dataset: Collected custom images + used online datasets.
Model Training: Trained a YOLOv8 model on the Mason Hopper Cluster.
Deployment: Deployed the model to a Raspberry Pi connected to a camera.
Hardware: Used servos and an Arduino to display a welcome message.
🧠 Challenges We Ran Into Difficulties in reliable communication between Raspberry Pi and Arduino (sending signals).
Mechanical challenges: initial design with a rotating platform using a stepper motor failed due to torque issues; had to redesign for stability and simplicity.
🏆 Accomplishments Successfully trained and deployed an accurate object detection model.
Built a real-world AI-powered sorting mechanism.
Integrated real-time LED feedback for users.
Overcame major technical and mechanical obstacles to deliver a working prototype.
📖 What We Learned Persistence matters — things rarely go exactly as planned.
Adaptability is crucial — sometimes redesigning is the best path forward.
Team collaboration — combining AI, hardware, and mechanical design required strong cross-functional teamwork.
🔮 What's Next Add multiple servo mechanisms to sort into three separate bins (bottles, cans, waste).
Refine the design for better durability and aesthetics.
Further fine-tune the model for real-world conditions.
Deploy SmartSort in public spaces like universities, airports, and parks.
🛠️ Built With Python
Raspberry Pi
Arduino
Roboflow
YOLOv8
🎯 Try It Out Want to build your own SmartSort? Here's how you can get started:
Requirements Raspberry Pi (any 4GB+ model recommended) Raspberry Pi Camera Module (or compatible USB camera) Arduino (Uno, Nano, or similar) Servo motors LED (optional for feedback) Jumper wires, breadboard, power supply 3D printed or cardboard frame (for building the sorting mechanism)
Installation
- Clone the Repository
git clone https://github.com/cheyennebejj123/smartsort.git
- Install Required Python Libraries
pip install -r requirements.txt
-
Set Up the Camera
-
Enable camera support on Raspberry Pi (raspi-config).
-
Test the camera using raspistill or a simple OpenCV script.
-
Upload Arduino Code and rasbperry PI
-
Run SmartSort
python smartsort.py
- Place an Item
Hold a plastic bottle, can, or trash item in front of the camera.
Watch SmartSort detect the item and automatically sort it into the correct bin