Seaway learning and motion compensation in shallow waters for small auvs

Abstract

The continual development of computer technology has enabled the expansion of intelligent control into the field of underwater robots, where potential uses include oceanographic research, environmental monitoring and military mine countermeasures. With the naval focus shifting to operations in the littorals, and the need to lower cost of operations, tetherless autonomous vehicles are now being proposed for use in very shallow water minefield reconnaissance. These areas are dominated by a highly energetic environment arising from waves and currents. Motion control in such an environment becomes a difficult task and is the subject of this work. The main objective of this dissertation, is to show through modeling and simulation, and in-ocean experimental validation, that intervention tasks performed by intelligent underwater robots are improved by their ability to gather, learn and use information about their working environment. Using a new generalized approach to the modeling of underwater vehicles, which directly includes disturbance effects, a new Disturbance Compensation Controller (DCC) is proposed. The DCC, employing onboard vehicle sensors, allows the robot to learn and estimate the seaway dynamics. This self-derived knowledge is embedded in a non-linear sliding mode control law which allows significantly improved motion stabilization. The performance of the DCC has been verified in Monterey Harbor using the NPS phoenix AUV