Smart Recycling Robot: Vision-Driven Autonomous Water Bottle Collection System using YOLOv5

Abstract

The growing concern for environmental sustainability has led to an increased focus on waste recycling and management. Many initiatives have been taken to promote waste recycling and garbage management. In this research, an idea of an autonomous water bottle collector is presented that combines computer vision to detect, approach, and pick up water bottles for recycling. This smart recycling robot can move around in the surroundings, detect the bottles lying on the ground, autonomously moving towards the detected bottle, and pick it. The system utilizes the YOLOv5 for object detection. It efficiently detects the bottles. Jetson Nano is used for processing the data and ESP 32 is used as a controller for the robot’s movement. Experimental results demonstrate the system’s ability to autonomously detect, approach, and pick up bottles with high accuracy and efficiency. The proposed water bottle collector robot offers a promising solution for automated recycling processes, reducing manual labour and promoting efficient waste.