A Study on Rule and Marine Environment Learning-based Unmanned Surface Vehicle Swarm Control

Abstract

The article describes a study on the development of a system for controlling multiple small unmanned surface vehicles (USVs) in a swarm for marine research purposes at sea. The study aimed to overcome limitations in acquiring broadband data in a wide range and diverse missions at sea by using swarm control techniques. The study used a long short-term memory (LSTM) model to learn disturbance information and predict maritime disturbances. The predicted ocean currents were used to generate a swarm USV control system for USV formations. The study conducted a comparative analysis of the designed USV model results and those generated by the simulator, and the effectiveness of the USV mathematical model and behavior control rules were verified. The system could potentially contribute to the exploration of marine data and resources