An integrated framework for autonomous near-real-time water quality monitoring using a solar-powered unmanned surface vehicles and artificial intelligence.

This paper presents the development of an unmanned solar vessel designed to collect and transmit water quality data in real time, together with a web-based system incorporating artificial intelligence (AI) for data processing, analysis, and visualization. The motivation for this development stems from the complexity and high costs associated with conventional water quality monitoring, particularly in remote, difficult-to-access, and large environments such as Amazonian hydropower reservoirs. The vessel was equipped with monitoring cameras, a single-beam echosounder to support navigation, and a multiparameter probe capable of profiling the water column. The AI module was developed to reduce reliance on laboratory analyses for the estimation of integrated water quality indicators, including the Water Quality Index (WQI). Field demonstrations conducted in the Intermediate Reservoir of the Belo Monte Hydroelectric Power Plant demonstrated the technical and operational feasibility of the proposed system. During the tests, the vessel autonomously navigated approximately 10 km in 3 h, successfully collected vertical water quality profiles at monitoring stations, and transmitted the data to the cloud, where the WQI was automatically estimated and made available through an interactive dashboard within a few minutes of data acquisition. By integrating autonomous sampling, cloud-based processing, AI-driven analysis, and near-real-time visualization into a single framework, the proposed system represents a promising complementary approach for water quality monitoring in remote aquatic environments. Future improvements and the integration of additional sensors (e.g., meteorological sensors) may further expand the monitoring capabilities of the platform, providing complementary information for water quality assessment and other environmental applications.