Online Ship Rolling Estimation Using a Grey Support Vector Machine Prediction Scheme

An online sequential grey predictive scheme is proposed by embedding the supporting vector machine (SVM) in a grey prediction framework. The grey processing of time series alleviates the unfavorable effects resulted from uncertainty available in the measurement data and the nonlinear and self-adaptation natures of SVM enable accurate approximation of the scheme. The resulted grey SVM predictor can be utilized to represent the nonlinear mapping influenced by uncertainty. The Ship’s motion at sea is affected by various time-varying environmental factors. As a result, the ship’s rolling motion is a complex nonlinear system which is hard to be predicted precisely by custom approaches. In this paper, the grey SVM predictor is utilized for online ship rolling angle prediction. The prediction simulation is performed based on the measurement data from scientific research and training ship Yu Kun. Simulation results have demonstrated that the proposed method can give predictions for ship rolling motion in real time with high accuracy and satisfactory stability.