—This paper predicts ship seakeeping based on nonlinear innovation. An improved extended Kalman filter (EKF) algorithm is proposed to predict ship pitch attitude based on Markov process model. The stochastic attenuation equation of coupled pitching motion is derived by using the Fokker-Planck equation. The parameters in these equations are then determined using the proposed algorithm. The hyperbolic tangent function is introduced and the nonlinear innovation method is used to process the error. By comparing the numerical simulation and full-scale test data, the effectiveness of the method is verified. The experimental data were obtained by using the “Yukun” ship. The method is more suitable for the case where the system is of small damping. Compared with traditional identification algorithms, the algorithm proposed in this paper has higher prediction accuracy. The real-time prediction results are compared with the full-scale data, which shows that the proposed ship prediction model has significant prediction accuracy and the algorithm is reliable. This parameter identification method can be used to establish ship maneuvering prediction model.