Safety-Guaranteed Ship Berthing Using Cascade Tube-Based Model Predictive Control

Abstract

This brief presents a cascade tube-based model predictive control (CTMPC) approach for safe ship berthing in the presence of unknown and bounded disturbances. The nonlinear dynamic model of a ship moving at low speeds is decomposed into a linear time-invariant (LTI) model for the surge system and a linear parameter-varying (LPV) model for the sway-yaw system, where the surge velocity is a scheduling parameter. The tube-based model predictive control (MPC) problems of these models are solved using the proposed CTMPC approach. The robust positively invariant (RPI) sets of the two subsystems and the geometry of the predefined separation area around the ship are considered to calculate the safe region of the ship under bounded disturbances. The proposed control architecture is shown to be both recursively feasible and stable, and simulation results are presented to verify its practical effectiveness.