Third M.I.T. Conference on Computational Fluid and Solid Mechanics June 14–17, 2005  

Real-time prediction of ship motion using artificial neural networks

A.A. Khana, C. Bila,*, K.E. Marionb
aRMIT University, School of Aerospace, Manufacturing and Mechanical Engineering, Melbourne, VIC 3061, Australia  bRMIT University, School of Mathematical and Geospatial Sciences, Melbourne, VIC 3000, Australia
  Full Text
ABSTRACT
Due to the random nature of ship motion in open water environments, ship-based deployment and recovery of aircraft can often be difficult and even dangerous. This paper presents an investigation into the application of artificial neural network methods for the prediction of ship motion relying only on measured ship data. It is shown that the artificial neural network produces excellent ship motion predictions and is able to predict the ship motion satisfactorily in real time for up to 10 seconds.

Keywords:  Ship motion; Artificial neural networks; Ship roll angle
* Corresponding author. Tel.: +61 3 9645 4537; Fax: +61 3 9645 4534; E-mail: cees.bil@rmit.edu.au