Nonlinear Innovation Identification for Norrbin Mathematical Model Via Actual Ship

In order to improve the prediction accuracy of ship maneuverability, this paper concentrates on the identification modeling problem on base of the ship trail data. Taking the Norrbin Mathematical model as an example, a nonlinear innovation identification algorithm is proposed by combine with the least squares theory and the concept of nonlinear innovation. In the algorithm, instead of using the original error information, the error innovation is derived by the nonlinear hyperbolic tangent function, which is selected to correct the related identification parameters. Based on the "Yukun" experimental ship, the trail test is carried out to obtain the data for the identification process. The comparison result has been illustrated to verify the advantages of the proposed algorithm, e.g., the high accuracy, the fast convergence speed and the clear physical meaning. It is obvious to note that the proposed algorithm can effectively solve the real-time identification problem for ship maneuverability and is with the guiding significance for the practical engineering.