A study on online hyper-heuristic learning for swarm robots

Shuang Yu, Andy Song, Aldeida Aleti

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

Abstract

Swarm robots continue to become more prominent in solving challenging tasks in real world applications. Due to the complexity of operating in often unknown environments, centralised control of swarm robots is not ideal. Prior manual programming is also not practical under these kind of circumstances. Thus, we establish a hyper-heuristic based learning approach for swarm robot control. With this framework, robots can autonomously identify appropriate heuristics from a set of given low-level heuristics, each heuristic guiding certain behaviours. We evaluated this type of online learning on building surface cleaning and studied the effectiveness of our hyper-heuristic online learning. Nine heuristics were proposed in this study. Through the experiments it can be seen that robots can improve their cleaning performance through the online learning process. More importantly, the experiments show that appropriate heuristics can be selected even when the size of the heuristic set is changed. The study on four types of environments shows that with the same heuristic set, the robot swarm can adapt to different environments for different tasks. Hence, hyper-heuristic learning is an effective method for decentralised control of swarm robots.

Original languageEnglish
Title of host publication2019 IEEE Congress on Evolutionary Computation (CEC) - 2019 Proceedings
EditorsMengjie Zhang, Kay Chen Tan
Place of PublicationPiscataway NJ USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages2721-2728
Number of pages8
ISBN (Electronic)9781728121536, 9781728121529
ISBN (Print)9781728121543
DOIs
Publication statusPublished - 2019
EventIEEE Congress on Evolutionary Computation 2019

- Wellington, New Zealand
Duration: 10 Jun 201913 Jun 2019
http://cec2019.org/

Conference

ConferenceIEEE Congress on Evolutionary Computation 2019

Abbreviated titleIEEE CEC 2019
CountryNew Zealand
CityWellington
Period10/06/1913/06/19
Internet address

Keywords

  • Hyper-heuristics
  • Online Learning
  • Robotic Surface Cleaner
  • Self-assembling Robots
  • Swarm Robots

Cite this

Yu, S., Song, A., & Aleti, A. (2019). A study on online hyper-heuristic learning for swarm robots. In M. Zhang, & K. Chen Tan (Eds.), 2019 IEEE Congress on Evolutionary Computation (CEC) - 2019 Proceedings (pp. 2721-2728). [8790164] Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/CEC.2019.8790164
Yu, Shuang ; Song, Andy ; Aleti, Aldeida. / A study on online hyper-heuristic learning for swarm robots. 2019 IEEE Congress on Evolutionary Computation (CEC) - 2019 Proceedings. editor / Mengjie Zhang ; Kay Chen Tan. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2019. pp. 2721-2728
@inproceedings{3f2de744b88c4cb2a6209a6951c14df7,
title = "A study on online hyper-heuristic learning for swarm robots",
abstract = "Swarm robots continue to become more prominent in solving challenging tasks in real world applications. Due to the complexity of operating in often unknown environments, centralised control of swarm robots is not ideal. Prior manual programming is also not practical under these kind of circumstances. Thus, we establish a hyper-heuristic based learning approach for swarm robot control. With this framework, robots can autonomously identify appropriate heuristics from a set of given low-level heuristics, each heuristic guiding certain behaviours. We evaluated this type of online learning on building surface cleaning and studied the effectiveness of our hyper-heuristic online learning. Nine heuristics were proposed in this study. Through the experiments it can be seen that robots can improve their cleaning performance through the online learning process. More importantly, the experiments show that appropriate heuristics can be selected even when the size of the heuristic set is changed. The study on four types of environments shows that with the same heuristic set, the robot swarm can adapt to different environments for different tasks. Hence, hyper-heuristic learning is an effective method for decentralised control of swarm robots.",
keywords = "Hyper-heuristics, Online Learning, Robotic Surface Cleaner, Self-assembling Robots, Swarm Robots",
author = "Shuang Yu and Andy Song and Aldeida Aleti",
year = "2019",
doi = "10.1109/CEC.2019.8790164",
language = "English",
isbn = "9781728121543",
pages = "2721--2728",
editor = "Zhang, {Mengjie } and {Chen Tan}, {Kay }",
booktitle = "2019 IEEE Congress on Evolutionary Computation (CEC) - 2019 Proceedings",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States of America",

}

Yu, S, Song, A & Aleti, A 2019, A study on online hyper-heuristic learning for swarm robots. in M Zhang & K Chen Tan (eds), 2019 IEEE Congress on Evolutionary Computation (CEC) - 2019 Proceedings., 8790164, IEEE, Institute of Electrical and Electronics Engineers, Piscataway NJ USA, pp. 2721-2728, IEEE Congress on Evolutionary Computation 2019

, Wellington, New Zealand, 10/06/19. https://doi.org/10.1109/CEC.2019.8790164

A study on online hyper-heuristic learning for swarm robots. / Yu, Shuang; Song, Andy; Aleti, Aldeida.

2019 IEEE Congress on Evolutionary Computation (CEC) - 2019 Proceedings. ed. / Mengjie Zhang; Kay Chen Tan. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2019. p. 2721-2728 8790164.

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

TY - GEN

T1 - A study on online hyper-heuristic learning for swarm robots

AU - Yu, Shuang

AU - Song, Andy

AU - Aleti, Aldeida

PY - 2019

Y1 - 2019

N2 - Swarm robots continue to become more prominent in solving challenging tasks in real world applications. Due to the complexity of operating in often unknown environments, centralised control of swarm robots is not ideal. Prior manual programming is also not practical under these kind of circumstances. Thus, we establish a hyper-heuristic based learning approach for swarm robot control. With this framework, robots can autonomously identify appropriate heuristics from a set of given low-level heuristics, each heuristic guiding certain behaviours. We evaluated this type of online learning on building surface cleaning and studied the effectiveness of our hyper-heuristic online learning. Nine heuristics were proposed in this study. Through the experiments it can be seen that robots can improve their cleaning performance through the online learning process. More importantly, the experiments show that appropriate heuristics can be selected even when the size of the heuristic set is changed. The study on four types of environments shows that with the same heuristic set, the robot swarm can adapt to different environments for different tasks. Hence, hyper-heuristic learning is an effective method for decentralised control of swarm robots.

AB - Swarm robots continue to become more prominent in solving challenging tasks in real world applications. Due to the complexity of operating in often unknown environments, centralised control of swarm robots is not ideal. Prior manual programming is also not practical under these kind of circumstances. Thus, we establish a hyper-heuristic based learning approach for swarm robot control. With this framework, robots can autonomously identify appropriate heuristics from a set of given low-level heuristics, each heuristic guiding certain behaviours. We evaluated this type of online learning on building surface cleaning and studied the effectiveness of our hyper-heuristic online learning. Nine heuristics were proposed in this study. Through the experiments it can be seen that robots can improve their cleaning performance through the online learning process. More importantly, the experiments show that appropriate heuristics can be selected even when the size of the heuristic set is changed. The study on four types of environments shows that with the same heuristic set, the robot swarm can adapt to different environments for different tasks. Hence, hyper-heuristic learning is an effective method for decentralised control of swarm robots.

KW - Hyper-heuristics

KW - Online Learning

KW - Robotic Surface Cleaner

KW - Self-assembling Robots

KW - Swarm Robots

UR - http://www.scopus.com/inward/record.url?scp=85071325172&partnerID=8YFLogxK

U2 - 10.1109/CEC.2019.8790164

DO - 10.1109/CEC.2019.8790164

M3 - Conference Paper

SN - 9781728121543

SP - 2721

EP - 2728

BT - 2019 IEEE Congress on Evolutionary Computation (CEC) - 2019 Proceedings

A2 - Zhang, Mengjie

A2 - Chen Tan, Kay

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway NJ USA

ER -

Yu S, Song A, Aleti A. A study on online hyper-heuristic learning for swarm robots. In Zhang M, Chen Tan K, editors, 2019 IEEE Congress on Evolutionary Computation (CEC) - 2019 Proceedings. Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. 2019. p. 2721-2728. 8790164 https://doi.org/10.1109/CEC.2019.8790164