Acquiring moving skills in robots with evolvable morphologies: Recent results and outlook

M. Jelisavcic; E. Haasdijk; A.E. Eiben

doi:10.1145/3067695.3084200

Acquiring moving skills in robots with evolvable morphologies: Recent results and outlook

M. Jelisavcic, E. Haasdijk, A.E. Eiben

Research output: Chapter in Book / Report / Conference proceeding › Conference contribution › Academic › peer-review

465 Downloads (Pure)

Abstract

We construct and investigate a strongly embodied evolutionary system, where not only the controllers but also the morphologies undergo evolution in an on-line fashion. In these studies, we have been using various types of robot morphologies and controller architectures in combination with several learning algorithms, e.g. evolutionary algorithms, reinforcement learning, simulated annealing, and HyperNEAT. This hands-on experience provides insights and helps us elaborate on interesting research directions for future development.

Original language	English
Title of host publication	GECCO '17
Subtitle of host publication	Proceedings of the Genetic and Evolutionary Computation Conference Companion
Publisher	ACM
Pages	1735-1741
Number of pages	7
ISBN (Electronic)	9781450349390
DOIs	https://doi.org/10.1145/3067695.3084200
Publication status	Published - Jul 2017

Keywords

Evolutionary robotics
Gait learning
Indirect encoding
Lamarckian evolution
On-line evolution

Access to Document

10.1145/3067695.3084200

jelisavcic2017acquiringAccepted author manuscript, 1.69 MB
Acquiring moving skills in robots with evolvable morphologiesFinal published version, 1.61 MBLicence: Article 25fa Dutch Copyright Act

Persistent URL (handle)

Copy link address

Cite this

@inproceedings{2730bb223f464cf6968251dd4cb7ecc1,

title = "Acquiring moving skills in robots with evolvable morphologies: Recent results and outlook",

abstract = "We construct and investigate a strongly embodied evolutionary system, where not only the controllers but also the morphologies undergo evolution in an on-line fashion. In these studies, we have been using various types of robot morphologies and controller architectures in combination with several learning algorithms, e.g. evolutionary algorithms, reinforcement learning, simulated annealing, and HyperNEAT. This hands-on experience provides insights and helps us elaborate on interesting research directions for future development.",

keywords = "Evolutionary robotics, Gait learning, Indirect encoding, Lamarckian evolution, On-line evolution",

author = "M. Jelisavcic and E. Haasdijk and A.E. Eiben",

year = "2017",

month = jul,

doi = "10.1145/3067695.3084200",

language = "English",

pages = "1735--1741",

booktitle = "GECCO '17",

publisher = "ACM",

}

Acquiring moving skills in robots with evolvable morphologies: Recent results and outlook. / Jelisavcic, M.; Haasdijk, E.; Eiben, A.E.
GECCO '17: Proceedings of the Genetic and Evolutionary Computation Conference Companion. ACM, 2017. p. 1735-1741.

Research output: Chapter in Book / Report / Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Acquiring moving skills in robots with evolvable morphologies: Recent results and outlook

AU - Jelisavcic, M.

AU - Haasdijk, E.

AU - Eiben, A.E.

PY - 2017/7

Y1 - 2017/7

N2 - We construct and investigate a strongly embodied evolutionary system, where not only the controllers but also the morphologies undergo evolution in an on-line fashion. In these studies, we have been using various types of robot morphologies and controller architectures in combination with several learning algorithms, e.g. evolutionary algorithms, reinforcement learning, simulated annealing, and HyperNEAT. This hands-on experience provides insights and helps us elaborate on interesting research directions for future development.

AB - We construct and investigate a strongly embodied evolutionary system, where not only the controllers but also the morphologies undergo evolution in an on-line fashion. In these studies, we have been using various types of robot morphologies and controller architectures in combination with several learning algorithms, e.g. evolutionary algorithms, reinforcement learning, simulated annealing, and HyperNEAT. This hands-on experience provides insights and helps us elaborate on interesting research directions for future development.

KW - Evolutionary robotics

KW - Gait learning

KW - Indirect encoding

KW - Lamarckian evolution

KW - On-line evolution

UR - http://www.scopus.com/inward/record.url?scp=85026890908&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85026890908&partnerID=8YFLogxK

UR - http://www.mendeley.com/research/acquiring-moving-skills-robots-evolvable-morphologies-recent-results-outlook

U2 - 10.1145/3067695.3084200

DO - 10.1145/3067695.3084200

M3 - Conference contribution

SP - 1735

EP - 1741

BT - GECCO '17

PB - ACM

ER -

Acquiring moving skills in robots with evolvable morphologies: Recent results and outlook

Abstract

Keywords

Access to Document

Persistent URL (handle)

Other files and links

Fingerprint

Cite this