A Generalized Dynamic Planning Framework for Green UAV-Assisted Intelligent Transportation System Infrastructure

Michael C. Lucic, Hakim Ghazzai, Yehia Massoud

Research output: Contribution to journalArticlepeer-review

17 Scopus citations


Roadside unit (RSU) planning is vital for the operation of an intelligent transportation system (ITS). RSUs provide ground coverage limited by obstacles. Unmanned aerial vehicles (UAVs) can complement RSU coverage by providing flexible connectivity capable of adapting coverage for traffic fluctuations, energy consumption, and budgetary constraints that all have effects on ITS operations. This article proposes a general RSU/UAV joint planning solution, where complex dynamic parameters are investigated. The objective is to maximize the effective coverage of placed RSUs and UAV docks given: a budget comprised of periodic operating expenses and capital expenditures, limitations of the ground transceivers and UAVs, and use of renewable energy to offset the on-grid electricity cost. We formulate a mixed-integer quadratically constrained problem that can determine the optimal placement of RSUs and UAV stations, RSU activation schedules, if solar panels are attached, and their coverage during each time period. Due to NP-hard complexity of such a planning problem, we design a heuristic algorithm that produces suboptimal solutions in less time. Afterward, we perform a sensitivity analysis and show that changes to the parameters lead to logical shifts in infrastructure coverage. Additionally, we visualize the algorithm's performance on a large setting - Manhattan Island.
Original languageEnglish (US)
Pages (from-to)4786-4797
Number of pages12
JournalIEEE Systems Journal
Issue number4
StatePublished - Dec 1 2020
Externally publishedYes

Bibliographical note

Generated from Scopus record by KAUST IRTS on 2022-09-13


Dive into the research topics of 'A Generalized Dynamic Planning Framework for Green UAV-Assisted Intelligent Transportation System Infrastructure'. Together they form a unique fingerprint.

Cite this