A Lower Bound for Controlled Lagrangian Particle Tracking Error

Abstract

Autonomous underwater vehicles are flexible mobile platforms for ocean sampling and surveillance missions. However, navigation of these vehicles in unstructured, highly variable ocean environments poses a significant challenge. Model-based prediction of vehicle position may be used to improve navigation capability, but prediction error exists due to limited resolution and accuracy of flow values obtained from ocean models that calculate flow velocity at discrete grid points. We present a theoretical lower bound on the steady-state error in position prediction for underwater vehicles using ocean model flow data and show that it is determined by the gridsize used by the ocean models. Our conclusions are justified by simulation and data collected during an ocean experiment.