Multivariable identification of ship steering and roll motions

This paper addresses identification of a multivariable model describing the rudder-induced surge-sway-yaw-roll dynamics of a fast container ship. The identification problem is formulated and is solved numerically by using a random search minimisation method Such method which has to carry out the estimation of a large number of unknown parameters, is based on a suitable implementation of Simulated Annealing. Application of the algorithm to a set of simulated data has proved to be quire successful. The results show that the proposed identification method is a very efficient and easy-to-use tool for determination of ship hydrodynamic derivatives on the basis of full-scale sea trials or from towing-lank tests conducted on scale models.