Continuous-time system identification of a ship on a river

Abstract

Model-based control strategies require accurate modeling of a system. Physical modeling leads to differential equations where the parameters can then be estimated from experimental data. In this paper, we present the continuoustime identification of the ship dynamics based on real data collected in open loop. In particular, the models for the drift and yaw dynamics are estimated for one ship. The obtained models show good results when tested with validation data and could be used, for example, for autopilot control strategies.

A Comparative and Experimental Study on Gradient and Genetic Optimization Algorithms for Parameter Identification of Linear MIMO Models of a Drilling Vessel

“…Bańka et al (Fossen, 1994) is then considered for two degrees of freedom of the ship's motion (sway and yaw), which after its linearization and adoption of certain simplifications is usually given in the form of transfer function models relating the yaw rate to the rudder angle and sway velocity to the rudder angle (Aström and Källström, 1976;1981;Velasco et al, 2013;Padilla and Yuz, 2013). Frequently, identification of ship models boils down to determining the hydrodynamic coefficients or hydrodynamic derivatives occurring in differential equations of the ship's model (Yoon and Rhee, 2003;Herrero and González, 2012;Zhang and Zou, 2011;Tran Khanh et al, 2013;Skjetne et al, 2004).…”