|
|
Publications
Moreira, L. and Guedes Soares, C. (2005), “Manoeuvring Simulation Based on Recurrent Neural Networks”, Proceedings of the 16th International Conference on Hydrodynamics in Ship Design - 3th International Symposium on Ship Manoeuvering, 7-10 September, Gdansk, Poland
A recurrent neural network based system is described and applied to surface ship manoeuvring simulation. The introduction of time process into the system is made based on the fact that the time component can be included through the introduction of a delay line. In this manoeuvring simulation approach the inputs to the system are the orders of rudder angle and ship’s speed and also the recursive outputs velocities of sway and yaw. Several results of the validation of the neural network model are presented: firstly using a manoeuvrability model performing the simulation of different manoeuvring tests for the Mariner hull to generate the data to train the network, and secondly using data obtained from full-scale trials to train the system. These results show that the recursive neural network system is a robust and accurate tool for manoeuvring simulation.
If you did not manage to obtain a copy of this paper: Request a copy of this article
For information about all CENTEC publications you can download: Download the Complete List of CENTEC Publications
|