A parameter estimation method for the seakeeping models of surface vehicles

The advent of autonomous marine systems has reshaped ocean observation, offering benefits such as reduced human exposure to harsh environments and enhanced spatiotemporal data collection. This paper presents a data-driven technique for estimating seakeeping models of small, unmanned surface vehicles (USVs) using motion measurements from onboard inertial sensors. The study highlights the limitations of traditional wave-to-motion transfer functions and the advantages of data-driven approaches in improving sea state estimation. A parameter estimation method is detailed, utilising closed-form expressions for response amplitude operators of simplified vessel models. The methodology is validated through a case study involving the NTNU AutoNaut USV, demonstrating accurate predictions of heave and pitch motions, although roll motion requires further refinement. The findings underscore the potential of data-driven methods in enhancing the fidelity of hydrodynamic models for USVs in complex wave environments.