Performance Comparison of PSO-based Informative Path Planners for Water Monitoring under Dynamic Scenarios

Abstract

The monitoring of water resources as an action of prevention and control of water quality is an ongoing case study. Monitoring can be performed using autonomous surface vehicles capable of measuring water quality parameters through in-vehicle sensors. This work focuses on comparing PSO-based informative path planners for water resource pollution peak detection with autonomous surface vehicles. The scenarios that are used for the comparison of the behavior of the algorithms are dynamic scenarios. In other words, the water resource contamination peaks are able to change position and size over time. The results show that the Enhanced GP-based PSO with exploring approach obtains the lowest error in detecting water resource contamination peaks with dynamic scenarios.