A data-driven method for optimal control of ship motions for safe crew transfer to offshore wind turbines

Abstract

Due to uncertainties and random behavior of sea loads, presenting an accurate hydrodynamic analysis for motion control of offshore ships is a big challenge. This paper aims to propose a novel method for motion control of a crew transfer vessel (CTV) in order to ensure safe crew transfer to an offshore wind turbine (OWT). For this purpose, we propose a novel neural network observer-based optimal control (NOPC) scheme to tackle unknown dynamics and disturbances, nonlinear effects, and non-symmetric control input saturation constraints. Accordingly, the neural network (NN) structure addresses the Hamilton-Jacobi-Bellman (HJB) equation and forms an optimal control signal remaining in the saturation bounds. The Lyapunov theory guarantees the Ultimately Uniformly Boundedness (UUB) of all signals of the closed-loop system. The high performance of the presented method is demonstrated in regular waves with high frequency in comparison with the previous studies. It is worth mentioning that there are not any limitations to implement the adopted strategy for other offshore applications.