SmartCan is a smart sorting bin designed to identify and collect recyclable materials using computer vision. Improper waste disposal and low recycling rates contribute to environmental issues, while individuals with mobility limitations may face challenges in properly disposing of waste. This project addresses both concerns by creating an autonomous system that can detect recyclable objects and move toward them. By combining accessibility with sustainability, SmartCan aims to promote cleaner environments and make recycling more convenient for a wider range of users.
SmartCan uses a camera paired with a real-time object detection model (YOLOv8) to identify recyclable items in its surroundings. The captured visual data is processed on a Raspberry Pi, which calculates the direction and distance of the detected object. This information is then transmitted to a Raspberry Pi Pico, which controls motor behavior. The Pico sends signals to L298N motor drivers, enabling precise movement of mecanum wheels to navigate toward the object. This system integrates computer vision, embedded systems, and motor control to achieve autonomous mobility.
By the end of the project, we developed a functional prototype of an autonomous trash-can robot capable of detecting and tracking recyclable objects using computer vision. The system successfully integrates hardware and software components, including sensors, a camera, microcontrollers, and motor drivers, into a cohesive embedded system with reliable communication.