PROJECT TITLE :

RESERVE: An Energy-Efficient Edge Cloud Architecture for Intelligent Multi-UAV

ABSTRACT:

Multi-unmanned aerial vehicle (MUAV) systems are able to perform tasks such as environmental and disaster monitoring, border surveillance, and search and rescue operations. This is due to the growing popularity of unmanned aerial vehicles (UAVs) in civil, public, and military applications. It is anticipated that applications based on multiple UAVs will become a significant pattern in edge computing scenarios in the near future. However, energy efficiency is a crucial concern in multi-UAV systems. This is because unmanned aerial vehicles (UAVs) have limited energy supplies, and they are also continually increasing the number of sensors they use. We believe that the combination of computing at the edge and computing in the cloud can provide effective support for the reduction of energy consumption. In this article, a cloud architecture for intelligent multi-UAV systems called RESERVE is presented that is both energy-efficient and environmentally friendly. Under the RESERVE program, we investigate a decentralized approach to solving the problem of energy-efficient computation offloading decision-making. The issue is presented in the form of a three-layer game, in which the discretionary strategy for achieving Nash equilibrium is detailed. In light of the game that has been suggested, we develop decentralized algorithms for two distinct use cases. Both of the algorithms are capable of reaching a Nash equilibrium state. In addition to this, we analyze the performance of the game based on its efficiency ratio and propose a mechanism for the decentralized offloading of computation. We run simulation experiments in addition to designing a prototype for the framework. The findings of the evaluation show that the proposed game methods have the potential to achieve a reduction in extra energy consumption of more than 30 percent in comparison to the most advanced decentralized algorithm and a reduction in performance loss of less than 10 percent in comparison to the centralized solution. The concept that we are advocating is demonstrated by the prototype framework that we have developed.