Development of the digital twin concept to optimize the interaction of robotic complexes in conditions of deep-sea mineral exploration

Abstract

Deep-sea mineral exploration is becoming a strategic direction for the resource development of the world's oceans, however, the extreme operating conditions of robotic complexes cause high risks and significant economic costs during operations. This study is devoted to the development of the concept of digital twins to optimize the interaction of robotic systems in a deep-sea environment. The proposed architecture of digital twins integrates a multi-level mathematical modeling system, simulation algorithms and machine learning technologies to predict the behavior of underwater complexes in real time. As part of the study, experimental testing of the developed concept was carried out on the basis of a sonar test site using three industrial autonomous uninhabited underwater vehicles. The test results demonstrate a 37.8% increase in the coordination efficiency of robotic systems and a 22.3% reduction in energy consumption compared to traditional control methods. The developed multi-agent optimization system has reduced the time required to perform technological operations by 41.5% while increasing the accuracy of underwater vehicle positioning by 18.9%. The presented concept forms a methodological basis for the development of intelligent control systems for robotic complexes in extreme conditions, contributing to increased economic efficiency and environmental safety of deep-sea mineral exploration.