Improved Model Structure for Ship Motion Identification Based on Reference Model and Bayesian Regularization Network

Previous artificial intelligence methods to system identification modeling for ship motion requires a mass of training data, modeling workload is vast. Aiming at these defects, an identification modeling method based on the reference model structure and Bayesian regularization network is proposed. For a start, an existed and public model is selected as the reference model. Secondly, With BR network, the reference model improves the generalization ability and reduces the training data. Finally, the method is verified with benchmark called KVLCC2. The illustrative example demonstrates the effectiveness and generalization ability of the proposed method.