An experimental study of adaptive bounded depth control for underwater vehicles subject to thruster’s dead-zone and saturation

Abstract

Accurate depth tracking control is of great importance for an underwater vehicle that performs an underwater task such as seabed mapping and underwater docking. This paper addresses the problem of adaptive depth tracking control for an underwater vehicle, particularly in the presence of hydrodynamics uncertainty, thruster’s dead-zone and saturation. Before deriving an adaptive depth control law, an unified and continuous hydrodynamics model in heave is built, where the thruster’s characteristics composed of output time-delay, input dead-zone and saturation are taken into consideration. Subsequently, fuzzy universal approximation theorem is resorted to online approximate the unknown and complex hydrodynamics in heave, and then adaptive sliding mode control technique is utilized to compensate for the approximation residual. In order to reject inherent input saturation, gradient projection method is introduced into the above adaptive control law, resulting in an adaptive bounded depth control law. Furthermore, through obtaining two dead-zone breakpoints in advance and online translation compensation, the final control output can always change within the permitted range. Finally, three comparative experiments are performed to verify the outperformance of the designed adaptive bounded depth tracking controller.