A separation principle for dynamic positioning of ships: theoretical and experimental results

TLDR

Ship state estimates are corrupted by white noise, so a passive observer is used to reconstruct the remaining states. The study presents a globally asymptotically stabilizing controller for ship regulation and dynamic positioning using only position measurements. The controller employs a passive observer that yields noise‑free estimates of position, disturbances, and velocity, which feed a PD‑type law; its stability is proved via a separation principle applicable to the nonlinear ship model. The separation principle is theoretically justified by cascaded nonlinear system results and Lyapunov theory, and experimentally validated on a 1:70 scale ship.

Abstract

Presents a globally asymptotically stabilizing (GAS) controller for regulation and dynamic positioning of ships, using only position measurements. It is assumed that these are corrupted with white noise hence a passive observer which reconstructs the rest of the states is applied. The observer produces noise-free estimates of the position, the slowly varying environmental disturbances and the velocity which are used in a proportional-derivative (PD)-type control law. The stability proof is based on a separation principle which holds for the nonlinear ship model. This separation principle is theoretically supported by results on cascaded nonlinear systems and standard Lyapunov theory, and it is validated in practice by experimentation with a model ship scale 1:70.