Search Space Analysis of Evolvable Robot Morphologies

K. da Silva Miras de Araujo, Evert Haasdijk, Kyrre Glette, A.E. Eiben

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

Abstract

We present a study on morphological traits of evolved modular robots. We note that the evolutionary search space –the set of obtainable morphologies– depends on the given representation and reproduction operators and we propose a framework to assess morphological traits in this search space regardless of a specific environment and/or task. To this end, we present eight quantifiable morphological descriptors and a generic novelty search algorithm to produce a diverse set of morphologies for any given representation. With this machinery, we perform a comparison between a direct encoding and a generative encoding. The results demonstrate that our framework permits to find a very diverse set of bodies, allowing a morphological diversity investigation. Furthermore, the analysis showed that despite the high levels of diversity, a bias to certain traits in the population was detected. Surprisingly, the two encoding methods showed no significant difference in the diversity levels of the evolved morphologies or their morphological traits.

Original languageEnglish
Title of host publicationApplications of Evolutionary Computation - 21st International Conference, EvoApplications 2018, Proceedings
EditorsKevin Sim, Paul Kaufmann
Place of PublicationCham
PublisherSpringer/Verlag
Pages703-718
Number of pages16
ISBN (Electronic)9783319775388
ISBN (Print)9783319775371
DOIs
Publication statusPublished - 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
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

Search Space
Encoding
Robot
Robots
Modular robots
Descriptors
Search Algorithm
Machinery
Operator
Demonstrate
Framework

Keywords

  • Evolutionary Robotics
  • Generative encoding
  • Modular robots
  • Morphology
  • Novelty search

Cite this

da Silva Miras de Araujo, K., Haasdijk, E., Glette, K., & Eiben, A. E. (2018). Search Space Analysis of Evolvable Robot Morphologies. In K. Sim, & P. Kaufmann (Eds.), Applications of Evolutionary Computation - 21st International Conference, EvoApplications 2018, Proceedings (pp. 703-718). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10784). Cham: Springer/Verlag. https://doi.org/10.1007/978-3-319-77538-8_47
da Silva Miras de Araujo, K. ; Haasdijk, Evert ; Glette, Kyrre ; Eiben, A.E. / Search Space Analysis of Evolvable Robot Morphologies. Applications of Evolutionary Computation - 21st International Conference, EvoApplications 2018, Proceedings. editor / Kevin Sim ; Paul Kaufmann. Cham : Springer/Verlag, 2018. pp. 703-718 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{d5e44549e5f34b37824fc63c39902239,
title = "Search Space Analysis of Evolvable Robot Morphologies",
abstract = "We present a study on morphological traits of evolved modular robots. We note that the evolutionary search space –the set of obtainable morphologies– depends on the given representation and reproduction operators and we propose a framework to assess morphological traits in this search space regardless of a specific environment and/or task. To this end, we present eight quantifiable morphological descriptors and a generic novelty search algorithm to produce a diverse set of morphologies for any given representation. With this machinery, we perform a comparison between a direct encoding and a generative encoding. The results demonstrate that our framework permits to find a very diverse set of bodies, allowing a morphological diversity investigation. Furthermore, the analysis showed that despite the high levels of diversity, a bias to certain traits in the population was detected. Surprisingly, the two encoding methods showed no significant difference in the diversity levels of the evolved morphologies or their morphological traits.",
keywords = "Evolutionary Robotics, Generative encoding, Modular robots, Morphology, Novelty search",
author = "{da Silva Miras de Araujo}, K. and Evert Haasdijk and Kyrre Glette and A.E. Eiben",
year = "2018",
month = "3",
doi = "10.1007/978-3-319-77538-8_47",
language = "English",
isbn = "9783319775371",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer/Verlag",
pages = "703--718",
editor = "Kevin Sim and Paul Kaufmann",
booktitle = "Applications of Evolutionary Computation - 21st International Conference, EvoApplications 2018, Proceedings",

}

da Silva Miras de Araujo, K, Haasdijk, E, Glette, K & Eiben, AE 2018, Search Space Analysis of Evolvable Robot Morphologies. in K Sim & P Kaufmann (eds), Applications of Evolutionary Computation - 21st International Conference, EvoApplications 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10784, Springer/Verlag, Cham, pp. 703-718, 21st International Conference on Applications of Evolutionary Computation, EvoApplications 2018, parma, Italy, 4/04/18. https://doi.org/10.1007/978-3-319-77538-8_47

Search Space Analysis of Evolvable Robot Morphologies. / da Silva Miras de Araujo, K.; Haasdijk, Evert; Glette, Kyrre; Eiben, A.E.

Applications of Evolutionary Computation - 21st International Conference, EvoApplications 2018, Proceedings. ed. / Kevin Sim; Paul Kaufmann. Cham : Springer/Verlag, 2018. p. 703-718 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10784).

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

TY - GEN

T1 - Search Space Analysis of Evolvable Robot Morphologies

AU - da Silva Miras de Araujo, K.

AU - Haasdijk, Evert

AU - Glette, Kyrre

AU - Eiben, A.E.

PY - 2018/3

Y1 - 2018/3

N2 - We present a study on morphological traits of evolved modular robots. We note that the evolutionary search space –the set of obtainable morphologies– depends on the given representation and reproduction operators and we propose a framework to assess morphological traits in this search space regardless of a specific environment and/or task. To this end, we present eight quantifiable morphological descriptors and a generic novelty search algorithm to produce a diverse set of morphologies for any given representation. With this machinery, we perform a comparison between a direct encoding and a generative encoding. The results demonstrate that our framework permits to find a very diverse set of bodies, allowing a morphological diversity investigation. Furthermore, the analysis showed that despite the high levels of diversity, a bias to certain traits in the population was detected. Surprisingly, the two encoding methods showed no significant difference in the diversity levels of the evolved morphologies or their morphological traits.

AB - We present a study on morphological traits of evolved modular robots. We note that the evolutionary search space –the set of obtainable morphologies– depends on the given representation and reproduction operators and we propose a framework to assess morphological traits in this search space regardless of a specific environment and/or task. To this end, we present eight quantifiable morphological descriptors and a generic novelty search algorithm to produce a diverse set of morphologies for any given representation. With this machinery, we perform a comparison between a direct encoding and a generative encoding. The results demonstrate that our framework permits to find a very diverse set of bodies, allowing a morphological diversity investigation. Furthermore, the analysis showed that despite the high levels of diversity, a bias to certain traits in the population was detected. Surprisingly, the two encoding methods showed no significant difference in the diversity levels of the evolved morphologies or their morphological traits.

KW - Evolutionary Robotics

KW - Generative encoding

KW - Modular robots

KW - Morphology

KW - Novelty search

UR - http://www.scopus.com/inward/record.url?scp=85044054660&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044054660&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-77538-8_47

DO - 10.1007/978-3-319-77538-8_47

M3 - Conference contribution

SN - 9783319775371

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

SP - 703

EP - 718

BT - Applications of Evolutionary Computation - 21st International Conference, EvoApplications 2018, Proceedings

A2 - Sim, Kevin

A2 - Kaufmann, Paul

PB - Springer/Verlag

CY - Cham

ER -

da Silva Miras de Araujo K, Haasdijk E, Glette K, Eiben AE. Search Space Analysis of Evolvable Robot Morphologies. In Sim K, Kaufmann P, editors, Applications of Evolutionary Computation - 21st International Conference, EvoApplications 2018, Proceedings. Cham: Springer/Verlag. 2018. p. 703-718. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-77538-8_47