Virtual Booth

This project presents the development of a low-cost underwater robotic system for autonomous navigation, coral monitoring, and acoustic-based surveillance in GPS-denied environments. The system is designed to address the limitations of conventional coral reef monitoring methods, which are often labour-intensive, costly, and constrained by environmental conditions. The proposed robot integrates a waterproof acrylic hull, propulsion system, pressure and inertial sensors, underwater camera, Raspberry Pi–based processing unit, and a custom graphical user interface for real-time control and monitoring. To enhance environmental perception, the system applies deep learning through YOLOv8 for real-time coral detection and classification, enabling intelligent observation of coral species and reef conditions. In addition, a vision-based 3D modelling approach is adopted to reconstruct coral reef structures without relying on expensive LiDAR equipment, making the system more affordable and scalable. The project combines mechanical design, embedded electronics, machine learning, and image-based reconstruction into a unified platform capable of underwater exploration, mapping, and ecological assessment. Overall, the developed robotic system demonstrates a practical, modular, and cost-effective solution for marine conservation, underwater research, and environmental monitoring applications.