Hydrodynamic Parameter Estimation for Autonomous Underwater Vehicles

This paper experimentally evaluates various hydrodynamic damping models for streamlined tail-controlled autonomous underwater vehicles (AUVs). We seek damping models that can be used to accurately predict the motion of a streamlined AUV, but do not possess unnecessary complexity. Additionally, we investigate the efficacy of estimating forces and moments due to control surfaces with the same data from which damping parameters are estimated. We then directly compare the results of estimated control surface forces and moments with that obtained using high-fidelity software tools. We use an offline least-squares method and an adaptive identifier method to approximate model coefficients from data attained from field trials. Field trials were conducted using two AUVs, and the experimental results are presented. The overall goal of this paper is to determine the form of a dynamic model that can be identified from data acquired during field trials.