EA-based ASV trajectory planner for detecting cyanobacterial blooms in freshwater

Abstract

Cyanobacterial Blooms (CBs) constitute a relevant ecological and public health problem since they often produce toxic metabolites that endanger the lives of many species, and they prevent human water consumption and recreational use. To determine the locations of CBs in lentic water bodies, we present a new planner based on Evolutionary Algorithms (EAs) that optimizes the trajectory of an Autonomous Surface Vehicle (ASV) equipped with a probe capable of detecting CBs. The planner 1) exploits the information provided by a particle transport simulator that determines the CB distribution from the water currents and the inherent CB behavior (in particular, its biological growth and vertical displacements) and 2) is supported by an EA that optimizes the mission duration, the ASV trajectory length, and the contributions of each simulated particle to the predicted cyanobacterial concentration along the ASV trajectory. The planner also ensures the trajectory feasibility from the ASV, probe, and water body perspective; and refines the trajectory shape by increasing the number of the decision variables during the iteration of an EA supported by usual NSGA-II operations. The results over different scenarios show that the planner determines overall good solutions that adapt the ASV trajectory to the evolution of CB distribution.