Jara Ten Kathen, M, Peralta, F, Johnson, P, Jurado Flores, I and Gutiérrez Reina, D (2024) AquaFeL-PSO: An informative path planning for water resources monitoring using autonomous surface vehicles based on multi-modal PSO and federated learning. Ocean Engineering, 311 (1). ISSN 0029-8018

Text AquaFeL_PSO_Ocean_MJ_R3_.pdf - Accepted Version Restricted to Repository staff only until 29 July 2026. Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (3MB)

Abstract

The preservation, monitoring, and control of large water resources has been a major challenge in recent decades. To ensure the quality of water resources, it is necessary to constantly monitor pollution levels. To meet this objective, this paper proposes an informative path planning for water resource monitoring, namely the AquaFeL-PSO algorithm, which uses autonomous surface vehicles, equipped with water quality sensors, based on a multi-modal particle swarm optimization, and the federated learning technique, with Gaussian process as a surrogate model. The proposed informative path planning has two phases, the exploration phase and the exploitation phase. In the exploration phase, the vehicles examine the surface of the water resource, and with the data acquired by the water quality sensors, an initial water quality model is estimated in the central server. In the exploitation phase, the area is divided into action zones using the model estimated in the exploration phase for a better exploitation of the contamination zones. To obtain the final water quality model of the water resource, the models obtained in both phases are combined. The results demonstrate that the proposed AquaFeL-PSO path planner is 3% more efficient in obtaining water quality models of the action zones than other similar path planners compared in this article. Moreover, in this case study, the AquaFeL-PSO has been demonstrated to have 300% more efficiency in achieving a better model of the entire water resource, and approximately 4,000% improvement in detecting pollution peaks. The analysis of variance provides support for these results, demonstrating a significant difference between the means of the path planners in detecting pollution peaks and generating the water quality model of the entire surface of the water resource. It was also shown that the results obtained by applying the federated learning technique are very similar to that of a centralized system and it allows to obtain a model of the water resource even when the Gaussian process of an action zone does not converge.

Item Type:	Article
Uncontrolled Keywords:	0405 Oceanography; 0905 Civil Engineering; 0911 Maritime Engineering; Civil Engineering
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) T Technology > TD Environmental technology. Sanitary engineering
Divisions:	Engineering
Publisher:	Elsevier
SWORD Depositor:	A Symplectic
Date Deposited:	20 Aug 2024 08:55
Last Modified:	20 Aug 2024 09:00
DOI or ID number:	10.1016/j.oceaneng.2024.118787
URI:	https://researchonline.ljmu.ac.uk/id/eprint/23910

View Item