Modified grey wolf optimizer-based support vector regression for ship maneuvering identification with full-scale trial

This study explores a nonparametric identification scheme for a ship maneuvering mathematical model. To overcome the difficulty in setting support vector regression (SVR) hyperparameters, a modified grey wolf optimizer algorithm is proposed. The algorithm introduces a nonlinear convergence factor and an adaptive position update strategy to enhance the search ability, which contributes toward identifying optimal hyperparameters. Using these optimal hyperparameters, SVR can predict the state variables pertaining to ship motion with high precision. The prediction of the motion state variables of the vessel YUKUN is considered as an illustrative example to verify the algorithm’s generalization ability and robustness. The prediction results indicate that, compared with the SVR based on the firefly algorithm and the particle swarm optimization, the proposed scheme offers the advantages of robustness, fewer iterations, and smaller prediction errors.