Hydrodynamic coefficient estimation for ship manoeuvring in shallow water using an optimal truncated LS-SVM

Abstract

An optimal truncated least square support vector machine (LS-SVM) is proposed for the parameter estimation of a nonlinear manoeuvring model in shallow water. A nonlinear manoeuvring model in shallow water is briefly introduced and the hydrodynamic coefficients are normalized. A series of Planar Motion Mechanism tests of the DTC ship model carried out in a towing tank with shallow water are used in the study for the estimation of the parameters of the manoeuvring model. The contribution of this paper is to propose the use of the optimal truncated LS-SVM, which is a robust method for parameter estimation that diminishes the uncertainty successfully and has a low computational cost by reducing the dimensionality of the kernel matrix. The classical least square method is also employed to estimate the parameters, and the results are compared with the optimal truncated LS-SVM. The parameter uncertainty is discussed, and the performance of the obtained nonlinear manoeuvring models is tested against a validation set that was left completely untouched during the training. The R2 goodness-of-fit criterion is used to demonstrate the accuracy of the obtained models.