Validation Test Report for a Genetic Algorithm in the Glider Observation STrategies (GOST 1.0) Project: Sensitivity Studies.

摘要

A suite of sensors in an oceanographic area of interest may be optimized with the use of a Genetic Algorithm (GA). The Environmental Measurements Path Planner (EMPath) executes a GA to generate optimal search plans for a suite of sensors based upon constituent cost-functions (CCF) contained in an input netcdf file. This GA software is incorporated to interface with the Relocatable Circulation Prediction System (RELO) under the Glider Observation Strategies (GOST) Project. This Validation Test Report (VTR) explores the sensitivity of a RELO to different Observation System Simulation Experiments (OSSEs) with simulated gliders in an area. Results are presented from a real-time exercise using the system, the Maritime Rapid Environmental Assessment of 2010 (MREA10), that include a full feedback cycle to guide a glider and assimilate the collected data back into the model.