Adaptive data transfer methods via policy evolution for UAV swarms

Phillip Smith, Robert Hunjed, Aldeida Aleti, Jan Carlo Barca

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

5 Citations (Scopus)


This paper presents an adaptive robotic swarm of Unmanned Aerial Vehicles (UAVs) enabling communications between separated non-swarm devices. The swarm nodes utilise machine learning and hyper-heuristic policy evolution to provide agility within the swarm, enabling each swarm member to select the most appropriate mobility policy for the environment given the swarm's abilities. The swarm evolution process of this study is found to successfully create different data transfer methods depending on the separation of non-swarm devices and the communication range of the swarm members. These methods are either human-designed, which the swarm adopts when most appropriate, or are novel hybridisations that the swarm creates for the problem. This paper also tests the swarm with individuals being removed during deployment. It is found that the swarm is immune to most alterations, though removal of specialised members of the heterogeneous swarm leads to temporary failure. The swarm evolution can then correct this failure by adjusting the swarm behaviour.

Original languageEnglish
Title of host publication2017 27th International Telecommunication Networks and Applications Conference (ITNAC 2017)
EditorsKhandakjar E. Ahmed, Keshav Sood
Place of PublicationPiscataway NJ USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Number of pages8
ISBN (Electronic)9781509067961, 9781509067954
ISBN (Print)9781509067978
Publication statusPublished - 2017
EventInternational Telecommunication Networks and Applications Conference 2017 - Melbourne, Australia
Duration: 22 Nov 201724 Nov 2017
Conference number: 27th (Conference website) (Proceedings)


ConferenceInternational Telecommunication Networks and Applications Conference 2017
Abbreviated titleITNAC 2017
Internet address


  • Hyper heuristic
  • Mobile ad hoc networks
  • mobility-assisted data delivery
  • reinforcement learning
  • Wireless communication

Cite this