On-line identification of the speed, steering and diving response parameters of an autonomous underwater vehicle from experimental data

Abstract

The experimental response data from autonomous maneuvering using the NPS AUV II vehicle has been analyzed with a view to defining Kalman filters to provide on-line estimates of system parameters and their variability. Kalman filters, designed for parameter estimation are expected to be the first step in the development of autonomous fault detection systems for underwater vehicles. Secondly, extraction of vehicle hydrodynamic coefficients from these parameters can help to develop vehicle dynamic simulators. Thirdly, knowledge of these parameters will allow the design of improved autopilot and guidance laws.