A Digital Twin Approach to Dynamic Obstacle Avoidance in UAV Swarm Operations

The idea of a “digital twin” has become a ground-breaking technological innovation in recent years, allowing the virtualization of physical systems to bridge the gap between the real and digital worlds. Meanwhile, Unmanned Aerial Vehicles (UAVs) are becoming increasingly integral to daily life. However, UAVs face constraints that hinder their optimal performance, such as limited battery life and processing capacity. Unlike UAVs, Digital Twins do not suffer from these limitations. In this paper, we propose a concept where UA V swarms and Digital Twins coexist. Specifically, we focus on collision avoidance, a critical aspect for the UAV swarms. Collision avoidance requires substantial computational effort and battery power. By offloading these computations to the Digital Twin, which can predict new formations to avoid collisions, we conserve UAV computational resources and battery life. We introduce a novel approach for formation prediction and test it on UAV swarms using edge computing and Digital Twins. Our results demonstrate that Digital Twins significantly reduce the processing load and communication requirements on the UAVs, thereby saving battery life and reducing collision incidents.