Low-cost collision avoidance in microverse for unmanned aerial vehicle delivery networks

Abstract

Unmanned Aerial Vehicles (UAV) have been widely adopted in many applications, from surveillance to delivery. More UAV delivery businesses are expected to be launched in the foreseeable future to meet food, goods, and medicine needs for residents living in smart cities, remote areas, or places lacking runways. As the density of UAVs operating in a community increases, collision avoidance becomes critical concerning the safety of personnel, property, and UAVs. In the last decade, many solutions have been suggested for collision avoidance scenarios, where typical solutions require integrated sensing, information exchange, and on-board decision-making. However, including these essential components increases the cost and makes it unaffordable for small-size UAVs in terms of payload weight and power consumption. Inspired by the Metaverse-enabled by Digital Twins, Blockchain, Augmented Reality (AR)/Virtual Reality (VR), and the fifth generation (5G) wireless communication technologies; we propose LoCASM, a low-cost collision avoidance scheme in Microverse, a local-scale Metaverse, for UAV delivery networks. LoCASM only requests position (GPS), altitude, velocity, and direction (PAVAD) information from each UAV; relieving the burden of expensive and energy-consuming components. By mirroring UAVs' PAVAD information and the city landscape in the Microverse, the computing-intensive tasks, including UAV tracking, trajectory prediction, and collision avoidance management, are migrated to the Microverse server on the ground. A proof-of-concept prototype of the LoCASM system has been built, and the simulation experimental study has validated the design.