Towards Multi-AUV Collaboration and Coordination: A Gesture-Based Multi-AUV Hierarchical Language and a Language Framework Comparison System

Abstract

The underwater environment is a harmful environment, yet one of the richest and least exploited. For these reasons the idea of a robotic companion with the task of supporting and monitoring divers during their activities and operations has been proposed. However, the idea of a platoon of robots at the diver’s disposal has never been fully addressed in these proposals due to the high cost of implementation and the usability, weight and bulk of the robots. Nevertheless, recent advancements in swarm robotics, materials engineering, deep learning, and the decreasing cost of autonomous underwater vehicles (AUVs), have rendered this concept increasingly viable. Therefore, this paper introduces, in the first part, a novel framework that integrates a revised version of a gesture-based language for underwater human–robot interaction (Caddian) based on insights gained from extensive field trials. The newly introduced objective of this framework is to enable the cooperation and coordination of an AUV team by one or more human operators, while allowing a human operator to delegate a robot leader to instruct the other robotic team members. The work, in the second part, provides an evaluation of the new language proposed thanks to a fifty million sentence corpus and describes a comparison framework, which is used to estimate it with respect to other existing underwater human–robot interaction languages.

An Efficient Multi-AUV Cooperative Navigation Method Based on Hierarchical Reinforcement Learning

“…Zhang et al [15] introduced an AUV cluster network navigation accuracy analysis method based on the Fisher information matrix and demonstrated that using the information on the entire formation when estimating the position of the following AUV could improve the positioning accuracy of the following AUV. Chiarella D. [16] designed a multi-AUV framework, defining a hierarchical work order of AUVs to efficiently complete cooperative tasks, and proposed a gesture-based UHRI framework to optimize coordination and communication between AUVs while supporting the other communication methods.…”