MANgO: federated world Model using an underwater Acoustic NetwOrk

Abstract

The usage of Autonomous Underwater Vehicles (AUVs) in scientific and industrial applications has increased in the last decade. In many applications the usage of fleet of heterogeneous assets is a key requirement. Their coordination and cooperation require high performance in terms of communication and information sharing. MANgO presents a federated world model operating over an underwater acoustic network through the networking capabilities provided by the SUNSET Software Defined Communication Stack (SDCS). It provides a method to transparently synchronise the local World Model (WM) of a vehicle with the rest of the heterogeneous team. The performance of the proposed architecture is demonstrated and compared with previous approaches using simulation and hardware-in-the-loop tests. Results show an improvement in the performance of at least 30% as well as the ease of integration with a variety of heterogeneous platforms.