An Aquaculture Big Data Platform for Unmanned Boats

As the aquaculture industry develops towards scale and refinement, the traditional management methods have been difficult to meet the modern aquaculture needs. This paper designs and develops an aquaculture big data platform based on unmanned boats, aiming to realize the intelligent management of water quality monitoring, underwater target recognition and cruise path planning. The hardware side is based on the selfdeveloped unmanned boat, which can collect water quality data such as water temperature, pH, TDS and underwater images in real time. The software side builds a visualization platform based on Vue and Leaflet framework, integrating Monte Carlo path planning, Kriging interpolation-fuzzy comprehensive evaluation and YOLOv5-UNet fusion recognition algorithms to achieve efficient data processing and analysis. The platform achieves millisecond transmission of massive data through EMQX, and supports real-time heat map display of water quality results, cruise path optimization and accurate identification of underwater targets. The experimental results show that the platform can significantly reduce labor costs and improve aquaculture efficiency, providing a feasible solution for the automation and intelligence of aquaculture.