Revolve: A Versatile Simulator for Online Robot Evolution

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

Abstract

Developing robotic systems that can evolve in real-time and real-space is a long term objective with technological as well as algorithmic milestones on the road. Technological prerequisites include advanced 3D-printing, automated assembly, and robust sensors and actuators. The necessary evolutionary mechanisms need not wait for these, they can be developed and investigated in simulations. In this paper, we present a system to simulate online evolution of constructible robots, where (1) the population members (robots) concurrently exist and evolve their morphologies and controllers, (2) all robots can be physically constructed. Experiments with this simulator provide us with insights into differences of using online and offline evolutionary setups.
LanguageEnglish
Title of host publicationApplications of Evolutionary Computation
Subtitle of host publicationEvoApplications 2018
EditorsKevin Sim, Paul Kaufmann
PublisherSpringer
Pages687-702
Number of pages16
Volume10784
ISBN (Electronic)9783319775388
ISBN (Print)9783319775371
DOIs
StateE-pub ahead of print - 8 Mar 2018
Event21st International Conference on Applications of Evolutionary Computation, EvoApplications 2018 - parma, Italy
Duration: 4 Apr 20186 Apr 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10784 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference21st International Conference on Applications of Evolutionary Computation, EvoApplications 2018
CountryItaly
Cityparma
Period4/04/186/04/18

Fingerprint

Simulator
Simulators
Robot
Robots
Constructible
Printing
Robotics
Actuator
Actuators
Real-time
Controller
Sensor
Controllers
Necessary
Sensors
Term
Experiment
Simulation
Experiments

Keywords

  • Evolutionary algorithms
  • Modular robots
  • Offline learning
  • Online learning
  • Reality gap

Cite this

Hupkes, E., Jelisavcic, M., & Eiben, A. E. (2018). Revolve: A Versatile Simulator for Online Robot Evolution. In K. Sim, & P. Kaufmann (Eds.), Applications of Evolutionary Computation: EvoApplications 2018 (Vol. 10784, pp. 687-702). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10784 LNCS). Springer. DOI: 10.1007/978-3-319-77538-8_46
Hupkes, Elte ; Jelisavcic, Milan ; Eiben, A. E./ Revolve: A Versatile Simulator for Online Robot Evolution. Applications of Evolutionary Computation: EvoApplications 2018. editor / Kevin Sim ; Paul Kaufmann. Vol. 10784 Springer, 2018. pp. 687-702 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{c4bc8ad8afc94bf099986c6bf5434977,
title = "Revolve: A Versatile Simulator for Online Robot Evolution",
abstract = "Developing robotic systems that can evolve in real-time and real-space is a long term objective with technological as well as algorithmic milestones on the road. Technological prerequisites include advanced 3D-printing, automated assembly, and robust sensors and actuators. The necessary evolutionary mechanisms need not wait for these, they can be developed and investigated in simulations. In this paper, we present a system to simulate online evolution of constructible robots, where (1) the population members (robots) concurrently exist and evolve their morphologies and controllers, (2) all robots can be physically constructed. Experiments with this simulator provide us with insights into differences of using online and offline evolutionary setups.",
keywords = "Evolutionary algorithms, Modular robots, Offline learning, Online learning, Reality gap",
author = "Elte Hupkes and Milan Jelisavcic and Eiben, {A. E.}",
year = "2018",
month = "3",
day = "8",
doi = "10.1007/978-3-319-77538-8_46",
language = "English",
isbn = "9783319775371",
volume = "10784",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer",
pages = "687--702",
editor = "Kevin Sim and Paul Kaufmann",
booktitle = "Applications of Evolutionary Computation",

}

Hupkes, E, Jelisavcic, M & Eiben, AE 2018, Revolve: A Versatile Simulator for Online Robot Evolution. in K Sim & P Kaufmann (eds), Applications of Evolutionary Computation: EvoApplications 2018. vol. 10784, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10784 LNCS, Springer, pp. 687-702, 21st International Conference on Applications of Evolutionary Computation, EvoApplications 2018, parma, Italy, 4/04/18. DOI: 10.1007/978-3-319-77538-8_46

Revolve: A Versatile Simulator for Online Robot Evolution. / Hupkes, Elte; Jelisavcic, Milan; Eiben, A. E.

Applications of Evolutionary Computation: EvoApplications 2018. ed. / Kevin Sim; Paul Kaufmann. Vol. 10784 Springer, 2018. p. 687-702 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10784 LNCS).

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

TY - GEN

T1 - Revolve: A Versatile Simulator for Online Robot Evolution

AU - Hupkes,Elte

AU - Jelisavcic,Milan

AU - Eiben,A. E.

PY - 2018/3/8

Y1 - 2018/3/8

N2 - Developing robotic systems that can evolve in real-time and real-space is a long term objective with technological as well as algorithmic milestones on the road. Technological prerequisites include advanced 3D-printing, automated assembly, and robust sensors and actuators. The necessary evolutionary mechanisms need not wait for these, they can be developed and investigated in simulations. In this paper, we present a system to simulate online evolution of constructible robots, where (1) the population members (robots) concurrently exist and evolve their morphologies and controllers, (2) all robots can be physically constructed. Experiments with this simulator provide us with insights into differences of using online and offline evolutionary setups.

AB - Developing robotic systems that can evolve in real-time and real-space is a long term objective with technological as well as algorithmic milestones on the road. Technological prerequisites include advanced 3D-printing, automated assembly, and robust sensors and actuators. The necessary evolutionary mechanisms need not wait for these, they can be developed and investigated in simulations. In this paper, we present a system to simulate online evolution of constructible robots, where (1) the population members (robots) concurrently exist and evolve their morphologies and controllers, (2) all robots can be physically constructed. Experiments with this simulator provide us with insights into differences of using online and offline evolutionary setups.

KW - Evolutionary algorithms

KW - Modular robots

KW - Offline learning

KW - Online learning

KW - Reality gap

UR - http://www.scopus.com/inward/record.url?scp=85044068273&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044068273&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-77538-8_46

DO - 10.1007/978-3-319-77538-8_46

M3 - Conference contribution

SN - 9783319775371

VL - 10784

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 687

EP - 702

BT - Applications of Evolutionary Computation

PB - Springer

ER -

Hupkes E, Jelisavcic M, Eiben AE. Revolve: A Versatile Simulator for Online Robot Evolution. In Sim K, Kaufmann P, editors, Applications of Evolutionary Computation: EvoApplications 2018. Vol. 10784. Springer. 2018. p. 687-702. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). Available from, DOI: 10.1007/978-3-319-77538-8_46