Model Based Predictive Control of Auvs for Station Keeping in a Shallow Water Wave Environment

An important consideration for Autonomous Underwater Vehicles in shallow water is station keeping. Station keeping is the ability of the vehicle to maintain position and orientation with regard to a reference object. In shallow water AUV operations, where large hydrodynamic forces are developed due to waves, knowledge of the sea is critical to allow for the design of a control system that will enable the vehicle to accurately navigate and position itself while in the presence of non-deterministic disturbances. Recently it has been shown that it is possible to measure the shape of the sea surface using remote sensors such as acoustic probes, lasers or short wavelength radar. These measurements have made it possible to develop so called "predictive" control strategies for many surface applications including hydrofoil operations and craning operations between vessels. The ability to develop predictive control strategies for underwater vehicles is limited by the ability to measure the environmental disturbances acting on the vehicle. In this paper we describe the design of a model based predictive controller that employs sub-surface sensors for disturbance prediction, to reduce wave induced effects on vehicle station keeping. Using linear wave theory and recursive estimation, an Auto Regressive (AR) model of the sea spectrum is developed. This dynamic model is then used to develop a forward predictor/estimator which is embodied in the controller to cancel the predictable portion of the non-deterministic disturbance on the vehicle thereby minimizing position error. Through simulation the ability of the Naval Postgraduate School's "PHOENIX" AUV to maintain longitudinal position while subjected to actual wave disturbance data from coastal Monterey Bay is shown.