Adaptive task allocation for multi-UAV systems based on bacteria foraging behaviour
Name
ASC_UTARB_revision_preprint.pdf
Description
Accepted version
Size
1.49 MB
Format
Unknown
Checksum (MD5)
ed2a84966eea11033449e31da19b5dbc
Author(s)
Kurdi, Heba
AlDaood, Munirah F.
Al-Megren, Shiroq
Aloboud, Ebtesam
Aldawood, Abdulrahman S.
Youcef-Toumi, Kamal
Date Issued
October 2019
Journal
Applied Soft Computing Journal
Publisher
Elsevier BV
Version
Author's final manuscript
Abstract
© 2019 Elsevier B.V. The foraging behaviour of bacteria in colonies exhibits motility patterns that are simple and reasoned by stimuli. Notwithstanding its simplicity, bacteria behaviour demonstrates a level of intelligence that can feasibly inspire the creation of solutions to address numerous optimisation problems. One such challenge is the optimal allocation of tasks across multiple unmanned aerial vehicles (multi-UAVs) to perform cooperative tasks for future autonomous systems. In light of this, this paper proposes a bacteria-inspired heuristic for the efficient distribution of tasks amongst deployed UAVs. The usage of multi-UAVs is a promising concept to combat the spread of the red palm weevil (RPW) in palm plantations. For that purpose, the proposed bacteria-inspired heuristic was utilised to resolve the multi-UAV task allocation problem when combating RPW infestation. The performance of the proposed algorithm was benchmarked in simulated detect-and-treat missions against three long-standing multi-UAV task allocation strategies, namely opportunistic task allocation, auction-based scheme, and the max-sum algorithm, and a recently introduced locust-inspired algorithm for the allocation of multi-UAVs. The experimental results demonstrated the superior performance of the proposed algorithm, as it substantially improved the net throughput and maintained a steady runtime performance under different scales of fleet sizes and number of infestations, thereby expressing the high flexibility, scalability, and sustainability of the proposed bacteria-inspired approach.
Subjects
Software
MIT Department
Massachusetts Institute of Technology. Department of Mechanical Engineering
Terms of Use
Creative Commons Attribution-NonCommercial-NoDerivs License
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1016/j.asoc.2019.105643
