Open AccessDynamic modelling of ship using gaussian processes and sg filter
Author(s) -
Gang Chen,
Wei Wang,
Huo Cong
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1861/1/012057
Subject(s) - noise (video) , gaussian process , nonlinear system , computer science , filter (signal processing) , control theory (sociology) , coupling (piping) , kalman filter , gaussian , engineering , artificial intelligence , control (management) , computer vision , physics , mechanical engineering , quantum mechanics , image (mathematics)
Dynamic modeling of surface ships is a prerequisite for intelligent navigation and ship motion control. It is easy to ignore the nonlinear and strong coupling of ship dynamics by using traditional mechanism modeling. Moreover, the obtained parameter errors will be large due to the influence of noise and multicollinearity, it is difficult to establish a high-precision ship dynamic model. This study use a data-driven nonparametric bayesian model based on gaussian process regression on dynamic modelling of ship, it can capture the strong nonlinearity and motion coupling in ship motion, and deal with the presence of uncertainty and noise. SG filter is used to smooth the data to reduce the influence of noise on modeling. The results indicate that the gaussian processes regression and the SG filter can reproduce the ship’s motion well, which is beneficial to the further development of ship control design.