“WAVECLEAN” – An Innovation in Autonomous Vessel Driving Using Object Tracking and Collection of Floating Debris

Marine pollution and floating marine debris are burgeoning concerns in the wake of industrial revolutions, posing a significant threat to aquatic ecosystems and marine-based economies. If not addressed promptly, this unchecked marine waste can lead to the demise of aquatic organisms, blockage of vital seaways, and the contamination of precious water resources. The disposal of materials such as plastics, glass, discarded fishing equipment, and industrial waste further disrupts the delicate chemical and biological balance of aquatic environments, endangering human health. Of all these pollutants, discarded plastics, which may eventually break down into microplastics, are particularly hazardous. Leveraging contemporary technology, the management of marine pollution has become more accessible. In this study, we focus on the autonomous driving component of a novel solution-an autonomous boat designed to collect floating trash from lakes, ponds, rivers, banks, and beaches. This autonomous boat is equipped with computer vision and deep learning technologies, employing models such as Mobile Net SSD, YOLOv5, and YOLOv8 in conjunction with OpenCV for real-time trash detection and identification. The boat operates autonomously, driven by a sophisticated system that employs Computer Vision and Convolutional Neural Networks (CNNs) for trash recognition. Video feed from an onboard camera aid in identifying floating debris, allowing the boat to navigate close to the trash for collection. Additionally, the boat can be controlled remotely or programmed to follow specific paths using the "ArduPilot" software, which offers mission planning tools for precise navigation. Beyond marine pollution mitigation, this autonomous boat serves a dual purpose by being readily adaptable for rescue operations during hydro-meteorological disasters. Equipped with a suite of sensors, it is well-suited for aquaculture applications.