Nonlinear MPC for Combined Motion Control and Thrust Allocation of Ships

Abstract For future autonomous marine vessels, better understanding of the ship’s behavior and control performance will be essential. Traditional motion control systems for ships decouple the problem into high-level motion control of the ship and thrust allocation to achieve the desired control action through the available actuators. The benefit is a segmented software, aiding in development and commissioning. The drawback of this decoupling is that the high-level controller at best has an approximate model of the capabilities in the thruster system. This typically leads to a mismatch between desired and achieved force especially when the control becomes aggressive. In this paper, a model predictive controller is proposed to solve both tasks simultaneously and overcome this drawback. The controller is based on a low-speed ship and thruster model and the resulting optimization problem is solved using the ACADO toolkit. A simulation study of a supply vessel with only two thrusters is presented to investigate the behavior of the proposed controller close to the boundary of low-speed maneuvering. The results show that there are benefits to incorporating the proposed controller.