A robust H-infinity based depth control of an autonomous underwater vehicle

This paper presents a robust H-infinity based control methodology for an Autonomous Underwater Vehicle(AUV). The kinematics and dynamics of an AUV is described using six degree of freedom differential equations of motion using body and earth-fixed frame of references. Due to hydrodynamic forces, these equations are highly coupled and non-linear. From the practical point of view it is essential to consider a reduced order model for efficient controller design. Hence the system is commonly subdivided into smaller subsystems, like depth, steering (or yawing) and speed subsystems, which are considered to be mutually non-interactive from the controller design perspective. In this study a reduced order model was derived using the depth plane dynamics of the vehicle. The working environment of an AUV is vastly uncertain due to varying environmental conditions, thereby demanding a robust controller which has the ability to adapt to these uncertainties and provide stabilizing effect irrespective of the change in the surrounding conditions. The proposed H-infinity controller takes into account the uncertainties in the hydrodynamic parameters which arise due to changing operating conditions and provides suitable control action for desired set point tracking as well as disturbance rejection. The altitude of the vehicle is strongly dependent on the pitch angle, and the controller presented here takes care of both the pitch and depth plane dynamics. The mixed sensitivity approach for H-infinity controller design is followed, and the efficacy of the controller is compared with Linear Quadratic Gaussian(LQG) controller and the Mixed H2/H-infinity controller. The controller design and simulation has been done in Matlab, and the simulated results provide satisfactory results, for disturbance rejection and set point tracking for the H-infinity controller in presence of hydrodynamic parametric uncertainties.