Digital Twin and Parallel Control for Automatic Berthing of Maritime Autonomous Surface Ships

The berthing of maritime autonomous surface ships (MASSs) is a difficult problem because the ships sail at low speed in ports with complex environment. This article investigates the automatic berthing control of an MASS subject to model uncertainties and ocean disturbances. An anti-disturbance parallel control architecture is proposed for the automatic berthing of an MASS. At the motion planning level, a digital twin model for the physical MASS is first established based on a neural-network state predictor method. Then, a model-predictive berthing planner is proposed to plan a smooth trajectory based on the digital twin model for automatic berthing. At the motion control level, a line-of-sight guidance law and a parallel anti- disturbance control law are proposed such that the MASS is able to track the planned berthing trajectory and achieve automatic berthing. It is proven that the closed-loop system is input-to-state stable. Simulation results are elaborated to substantiate the efficacy of the proposed parallel berthing control architecture for an MASS.