Airborne Computing: A Toolkit for UAV-Assisted Federated Computing for Sustainable Smart Cities

Document Type

Article

Source of Publication

IEEE Internet of Things Journal

Publication Date

1-1-2023

Abstract

Smart vehicles are equipped with onboard computing units designed to run in-vehicle applications. However, due to limited computing power, the onboard units are unable to execute compute-intensive tasks and those that require near real-time processing. Therefore tasks are offloaded to nearby fog/ edge devices that have more powerful processors. However, the fog devices are static, placed at fixed locations such as intersections, and have a limited communication range. Therefore they can only facilitate vehicles in their immediate vicinity and only limited areas of the city can be covered to provide services on demand. In this paper, we propose a UAV-based computing framework design termed Skywalker to provide computing in regions where there are no static fog units thereby extending coverage. Skywalker’s contributions are three-fold: (1) It allows for load-aware UAV placement and provisions a swarm of UAVs to fly to areas experiencing a gap in service where the size of the swarm is proportional to the demand. (2) It implements multiple scheduling algorithms that the UAVs swarm employs to divide up the task processing responsibility for individual UAVs within the swarm. (3) A zone-based delivery mechanism is being proposed to facilitate the return of completed tasks, either through direct delivery or relay-based methods. The choice between these options depends on the distance covered by the requesting vehicle from the UAV swarm. The efficiency of the framework is compared with existing techniques and it is found that it can greatly extend coverage during peak traffic hours while providing low communication delay and consuming minimum energy.

ISSN

2327-4662

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Disciplines

Computer Sciences

Keywords

Autonomous aerial vehicles, Edge computing, fog computing, Real-time systems, Relays, Servers, smart cities, Smart cities, Task analysis, task offloading, UAV swarm

Scopus ID

85164421301

Indexed in Scopus

yes

Open Access

no

Link to Full Text

Share

COinS