Algorithm for Controlling an Autonomous Underwater Vehicle when Avoiding a Complex Obstacle

The problem of avoiding obstacles by an autonomous underwater vehicle is considered. In contrast to the traditional approach to choosing a route around an obstacle, based on its current observation, the work proposes to additionally use an electronic bathymetric map and a physical map of the world, as well as the results of aerial photography - to obtain complete information about the position, shape and size of the obstacle. This approach allows you to reasonably select the shortest route around a detected obstacle. Due to the fact that the depth values are indicated in grid nodes with an unacceptably large step, to increase the detail of the bottom topography, the Kriging algorithm was used in the work, which makes it possible to refine the depth grid. The operation of the Kriging algorithm is clearly demonstrated, and an algorithm for determining the spatial shortest route is developed. Using a specially developed mathematical model, numerical experiments were carried out aimed at testing the proposed solution to bypass the obstacle and assess the gains from the use of additional information in the form of digital electronic maps. the results of a numerical experiment showed that the use of additional materials led to a reduction in the bypass path by 10-17%, which leads, accordingly, to a reduction in the time to bypass an obstacle and save energy resources.