Intelligent vector-based path following guidance law for unmanned surface vehicles

This study proposes a significant improvement on the classical vector field guidance law. The classical vector field method results in an imaginary number in some cases. In the numerical calculation, the imaginary part of the number is ignored in order to obtain a solution. Neglecting the imaginary part adversely impacts the optimization process, resulting in increased cross tracking errors. The first objective of this study is to modify the classical vector field method in order to obtain real numbers in all scenarios. Owing to this modification, cross tracking errors were reduced by 10% compared to the traditional vector field method. The second goal of this study is to incorporate the surrogate optimization technique into the guidance laws developed in this study (the modified vector field and the robust vector field) to quickly tune the parameters of these guidance laws. The classical vector field and other algorithm improved in this study require optimizing four key parameters. While the unmanned surface vehicle performs the multi-command tasks, determining the optimum values of these four key parameters for each mission increases the computational costs significantly. The proposed intelligent vector-based path following guidance law (which integrates the surrogate optimization technique with the robust vector field) determines optimal parameters approximately 10 times faster than the traditional method (which integrates the genetic algorithm with the classical vector field guidance law).