Effects of selection preferences on evolved robot morphologies and behaviors

Karine Miras*, Evert Haasdijk, Kyrre Glette, A. E. Eiben

*Corresponding author for this work

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

Abstract

This paper investigates the evolution of modular robots using different selection preferences (i.e., fitness functions), aiming at novelty, speed of locomotion, number of limbs, and combinations of these. The outcomes are analyzed from different perspectives: sampling of the search space, evolved morphologies, and evolved behaviors. This results in a wealth of findings, including a surprise about the number of sampled regions of the search space and the effect of different fitness functions on the evolved morphologies.

Original languageEnglish
Title of host publicationProceedings - ALIFE 2018
Subtitle of host publicationThe 2018 Conference on Artificial Life
EditorsTakashi Ikegami, Nathaniel Virgo, Olaf Witkowski, Mizuki Oka, Reiji Suzuki, Hiroyuki Iizuka
PublisherMassachusetts Institute of Technology
Pages224-231
Number of pages8
DOIs
Publication statusPublished - Jul 2018
Event2018 Conference on Artificial Life: Beyond AI, ALIFE 2018 - Tokyo, Japan
Duration: 23 Jul 201827 Jul 2018

Publication series

NameArtificial Life Conference Proceedings
PublisherMIT
Volume30

Conference

Conference2018 Conference on Artificial Life: Beyond AI, ALIFE 2018
Country/TerritoryJapan
CityTokyo
Period23/07/1827/07/18

Keywords

  • Evolutionary robotics
  • Generative encoding
  • Morphology evolution
  • Novelty search
  • Search space

Fingerprint

Dive into the research topics of 'Effects of selection preferences on evolved robot morphologies and behaviors'. Together they form a unique fingerprint.

Cite this