Estimation of Ship Oscillation subject to Ship Speed Variation Using an Algorithm Combining H∞ and Kalman Filters

This paper presents an estimation method of ship oscillation subject to ship speed variation, which leads to oscillation frequency variation. Such variation will deteriorate the estimation performance particularly when relying on a model-based estimation algorithm such as the Kalman filter. To overcome the problem, we propose an estimation method of ship oscillation with a low-cost gyro rate sensor, which incorporates periodic updates of fast Fourier transform (FFT)-based model and a method of selectively combining multiple H-infinity and Kalman filters according to an innovation-based criterion. The latter algorithm is the heart of our estimation method, which aims at utilizing the advantage of each filter. Some simulation results show that the proposed method is more effective than the conventional H-infinity and Kalman filters.