Development of a path planning algorithm for a water unmanned surface vehicle using particle swarm optimization method for water quality monitoring

Abstract Water pollution driven by industrial activities and human neglect necessitates advanced, real-time monitoring methods. Traditional sampling techniques are labor-intensive and lack immediacy in data acquisition. This study presents an Unmanned Surface Vehicle (USV) system employing a Particle Swarm Optimization (PSO) algorithm for efficient path planning, reducing navigation time and enhancing battery efficiency. Integrating autonomous navigation, IoT capabilities, and pH and electrical conductivity sensors, the USV enables accurate water quality evaluations with minimal human intervention. Simulations using datasets such as Oliver30, Eil51, and St70 demonstrated competitive performance of the modified PSO, with an average convergence time of 27 seconds and a 16% error rate across the datasets. Physical tests validated the system’s ability to traverse water, yielding precise water quality readings and localizing data with a variance of ≈11-16 cm. Limitations in Ultra-wideband localization for outdoor environments suggest the potential for GPS integration and expanded sensor capabilities in future work. This research highlights the transformative potential of USVs for real-time environmental monitoring, offering scalable solutions for managing and preserving aquatic ecosystems.