Edge Computing Enabled by Unmanned Autonomous Vehicles

TL;DR

This paper introduces GEESE, a system where UAVs transport cloudlets to provide dynamic edge computing support, reducing energy footprints and enhancing pervasive application performance in various environments.

Contribution

The paper presents GEESE, a novel UAV-based system for deploying mobile edge computing infrastructure through cooperative UAVs carrying cloudlets.

Findings

UAVs can effectively deliver cloudlets in diverse environments.

Cooperative UAVs improve the efficiency of edge computing deployment.

Experimental results show feasibility of UAV-based edge computing support.

Abstract

Pervasive applications are revolutionizing the perception that users have towards the environment. Indeed, pervasive applications perform resource intensive computations over large amounts of stream sensor data collected from multiple sources. This allows applications to provide richer and deep insights into the natural characteristics that govern everything that surrounds us. A key limitation of these applications is that they have high energy footprints, which in turn hampers the quality of experience of users. While cloud and edge computing solutions can be applied to alleviate the problem, these solutions are hard to adopt in existing architecture and far from become ubiquitous. Fortunately, cloudlets are becoming portable enough, such that they can be transported and integrated into any environment easily and dynamically. In this article, we investigate how cloudlets can be transported by unmanned autonomous vehicles (UAV)s to provide computation support on the edge. Based on our study, we develop GEESE, a novel UAVbased system that enables the dynamic deployment of an edge computing infrastructure through the cooperation of multiple UAVs carrying cloudlets. By using GEESE, we conduct rigorous experiments to analyze the effort to deliver cloudlets using aerial, ground, and underwater UAVs. Our results indicate that UAVs can work in a cooperative manner to enable edge computing in the wild.