Online short-term ship response prediction with dynamic buffer window using transient free switching filter

Considering the broad applications in autonomous marine vehicle control, ship response prediction has emerged as a significant area of interest in seakeeping. In particular, the short-term prediction and online updates of the ship response have been in the practical attention span. Although numerous studies have explored various predictors, the asymptotic properties of estimators employing different window lengths of time series have been neglected. Existing analyses focus on fixed observation windows without considering intrinsic statistical traits within the time series. The current research has investigated the performance of two nonlinear and linear based regressors: support vector regression (SVR) and the adaptive Seasonal Auto-Regressive and Integrated Moving Average (SARIMA) models on the data obtained from a simulated semisubmersible platform in different sea states. Results show prediction accuracy's strong dependence on observation window length relative to response statistical properties under wave conditions. Therefore, an innovative nonlinear filter has been introduced to dynamically adjust the buffer window size based on the signal statistical characteristics. The promising results showed that the filter not only could contribute to promoting any predictors used in autonomous onboard decision-making and stabilization systems for intelligent onboard control but the proposed mechanism can be employed in dynamic systems in other disciplines.