A two-layered fast marching path planning algorithm for an unmanned surface vehicle operating in a dynamic environment

Abstract

This paper describes a modified fast marching (FM) path planning algorithm for unmanned surface vehicles (USVs). The modified FM algorithm generates a two-layered synthetic vector field to represent a dynamic environment. The synthetic vector field integrates the obstacle information and the environment information, where current and wind velocities vary in both magnitude and direction. The path planning algorithm then employs the anisotropic FM method to calculate a safe trajectory to avoid obstacles and to minimise any negative effects of the environment. The algorithm has been tested in the environment with simulated current. The resulting trajectory shows that the two-layered FM algorithm is able to deal with environmental influence satisfactory.