An application of model predictive control for a shipboard crane

Designing an effective controller for shipboard cranes by means of analytical methods is too complex, because of their highly nonlinear equations of motion. In this work we formulate the swinging suppression problem for a new structure of shipboard crane designed by Maryland research group. Then a feedforward control law to greatly decrease the load sway will be introduced. This feedforward control counteracts the effect of ship rolling on load sway based on measurements of ship rolling angle at each instant. The measurement errors and also sways caused by other disturbance sources are not taken into account by this feedforward control. Next, we describe the model predictive control (MPC) based nonlinear controller that is designed for this problem. The proposed controller uses an optimizer to find an open loop solution at each sampling interval for a given horizon, based on a model of the plant and an appropriately defined objective function. MPC acts as a feedback control that will compensate for shortcomings of the feedforward control. Details of the controller, the design process and the simulation results are presented