Learning control for underwater robotic vehicles

Abstract

Underwater robotic vehicles have become an important tool for various underwater tasks because they have greater speed, endurance, and depth capability, as well as a higher factor of safety, than human divers. However, most vehicle control system designs have been based on a simplified vehicle model, which has often resulted in poor performance because the nonlinear and time-varying vehicle dynamics have parameter uncertainty. It was also observed by experiment that the thruster system had nonlinear behavior and its effect on vehicle motion was significant. It is desirable to have an advanced control system with the capability of learning and adapting to changes in the vehicle dynamics and parameters. This article describes a learning control system using neural networks for underwater robotic vehicles. Its effectiveness is investigated by simulation.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>