Título Monitoring Peak Pollution Points of Water Resources with Autonomous Surface Vehicles Using a PSO-Based Informative Path Planner
Autores JARA TEN KATHEN, MICAELA, Johnson P. , JURADO FLORES, ISABEL, Gutiérrez Reina D.
Publicación externa No
Medio Studies in Computational Intelligence
Alcance Capítulo de un Libro
Naturaleza Científica
Cuartil SJR 4
Impacto SJR 0.208
Web https://www.scopus.com/inward/record.uri?eid=2-s2.0-85163823666&doi=10.1007%2f978-3-031-26564-8_4&partnerID=40&md5=fc39f13c2de486e8f85622078f7501f5
Fecha de publicacion 01/01/2023
Scopus Id 2-s2.0-85163823666
DOI 10.1007/978-3-031-26564-8_4
Abstract The preservation of water resources is an increasingly urgent issue. Therefore, monitoring the water quality of these resources is a very important task so that appropriate actions could be taken. This chapter focuses on water resource monitoring using a fleet of Autonomous Surface Vehicles equipped with sensors capable of measuring water quality parameters. The objective is to obtain the maximum points of contamination of the water resource through the exploration and exploitation of the water surface. The proposed algorithm is based on Particle Swarm Optimization (PSO) in combination with some machine learning techniques (Gaussian Process, Bayesian Optimization, among others) to address the limitations of PSO, such as premature convergence and difficulty in setting the initial values of the coefficients. To validate the performance of the algorithm, uni-modal and multi-modal benchmark functions are used in the simulation experiments. The results show that the proposed algorithm, the Enhanced GP-based PSO, based on the epsilon greedy method has the best performance for detecting water resource pollution peaks. It was also demonstrated that this algorithm is the one that generates the most accurate water quality model. However, when it comes to finding the highest pollution peak, the algorithm with the best response is the Enhanced GP-based PSO with a focus on exploitation. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
Palabras clave Autonomous surface vehicles; Informative path planning; Machine learning; Multi-modal problems; Particle swarm optimization; Water monitoring
Miembros de la Universidad Loyola

Change your preferences Gestionar cookies