Intelligent Beach Cleaning Robot with Dual Modes of Refuse Collection and YOLO-based Detection

Abstract

Coastal pollution, driven by the accumulation of plastics and debris on beaches, poses a serious threat to natural habitats and marine ecosystems. Manual cleaning efforts, while valiant, are labor intensive and insufficient to address the scale of the problem. Intelligent beach cleaning robots offer a promising solution by integrating advanced technology with environmental stewardship. This paper reviews existing robotic refuse collection mechanisms and introduces an Intelligent Beach Cleaning Robot equipped with dual modes of refuse collection mechanisms (raking and sieving shaker), object detection, autonomous navigation, and obstacle avoidance systems. Through rigorous testing and analysis, we achieved an accuracy of $93 \%$ in plastic debris detection using the YOLOv5 model. The robot’s raking mechanism effectively collects debris within a 6 -second time frame, while the sieving shaker mechanism operates at an optimal frequency of 2.5 Hz for efficient sand filtration. The proposed prototype aims to improve the refuse collection capabilities cost-effectively, addressing the critical need for adaptable and efficient beach cleaning solutions.