PSO method for optimizing loss function weights for ship model control

The article presents the development of the particle swarm optimization (PSO) method and its application in the interpretation of land data and in maritime contexts, including navigation and determining the safe trajectory of a surface vessel. In particular, the article explores how the PSO algorithm can be used to optimize the trajectory of the “Blue Lady” ship model. The author decided to present the problem from a different point of view. The author adopts a novel perspective by first presenting a maneuver conducted in the port of the Foundation for Safety of Navigation and Environmental Protection, where both a linear matrix inequality (LMI) controller and a trajectory master controller were employed. The PSO method was subsequently applied to minimize the maneuver execution time by optimizing the weighting parameters of the objective function. This optimization considered the minimization of the https://www.w3.org/1998/Math/MathML" display="inline"> H ∞ norm and the https://www.w3.org/1998/Math/MathML" display="inline"> H 2 norm, and promising results were obtained. The innovative use of the PSO method ultimately demonstrated that the low-speed trajectory of the training vessel was completed in a shorter time compared to the prototype maneuver (performed on a lake), which did not incorporate the PSO method.