Multi-vehicle autonomous sampling of a coastal thermal and effluent jet and plume

Abstract

Adaptive sampling algorithms and behavior-based approaches can aid in the rapid and accurate in-situ measurement and characterization of coastal environmental features such as industrial thermal effluent jets and plumes. To enable the development of these techniques we present a collection of simulation, estimation, and field tools for use within the Mission Oriented Operations Suite (MOOS). Key features include a multiparameter model of thermal effluent jets and plumes, simulated annealing parameter estimation, and a multi-sensor indicator function. Using these tools, an adaptive multi-vehicle transect sampling behavior is implemented to efficiently sample an industrial jet. The capabilities of this behavior are demonstrated in realistic mission simulations and in field trials using a fleet of autonomous surface vehicles.