Global dynamic window approach for autonomous underwater vehicle navigation in 3d space

Abstract

The marine world becomes more narrow and full of different objects that move unpredictably in the ocean space. The problem of increasing the capacity of the systems management in any kind of underwater robots is highly relevant based on the development of new methods for the dynamic analysis, pattern recognition, artificial intelligence and adaptation. Among the huge number of navigation methods, Dynamic Window Approach is worth noting. It was originally presented by Fox et al. and implemented into indoor office robots. In this paper Dynamic Window Approach was developed for marine world and extended to manipulate the vehicle in 3D environment. This algorithm is provided to avoid obstacles and reach targets in efficient way. It was tested using MATLAB environment and assessed as an effective obstacle avoidance approach for marine vehicles.