Adaptive Weighted Least Squares Support Vector Regression Based on Genetic Algorithm for Ship Maneuvering Model Identification

Abstract

In order to improve the accuracy of least squares support vector regression (LS-SVR) in parameter identification of ship maneuvering model, adaptive weighted least squares support vector regression machine (AWLS-SVR) with sparsity and robustness is selected as the basic identification method, and genetic algorithm (GA) is used to optimize its structural parameters. The AWLS-SVR based on genetic algorithm proposed in this paper has been verified in the maneuver experimental research on a self-built surface unmanned vessel (USV). Furthermore, the identification results of experimental data are used to determine a more appropriate ship maneuvering motion model to make it more consistent with the maneuverability of the experimental ship. The experimental results indicate that the proposed hybrid intelligent identification method has higher identification accuracy than the AWLS-SVR.