EUVIMEDEUVIMED entdecken
Vergleichsstudie

Comparing Robot Controller Optimization Methods on Evolvable Morphologies

Fuda van Diggelen iD, Eliseo Ferrante iD, A E Eiben iD

Evolutionary Computation · 2024

DOI: 10.1162/evco_a_00334 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift Evolutionary Computation, 2024. Verfasst von Fuda van Diggelen und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, Vrije Universiteit Amsterdam, The Netherlands fuda.van.diggelen@vu.nl. Hauptthemen laut MEDLINE: Robotics, Algorithms, Bayes Theorem. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 37 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In this paper, we compare Bayesian Optimization, Differential Evolution, and an Evolution Strategy employed as a gait-learning algorithm in modular robots. The motivational scenario is the joint evolution of morphologies and controllers, where “newborn” robots also undergo a learning process to optimize their inherited controllers (without changing their bodies). This context raises the question: How do gait-learning algorithms compare when applied to various morphologies that are not known in advance (and thus need to be treated as without priors)? To answer this question, we use a test suite of twenty different robot morphologies to evaluate our gait-learners and compare their efficiency, efficacy, and sensitivity to morphological differences. The results indicate that Bayesian Optimization and Differential Evolution deliver the same solution quality (walking speed for the robot) with fewer evaluations than the Evolution Strategy. Furthermore, the Evolution Strategy is more sensitive for morphological differences (its efficacy varies more between different morphologies) and is more subject to luck (repeated runs on the same morphology show greater variance in the outcomes).

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

RoboticsAlgorithmsBayes TheoremGaitHumansComputer SimulationArtificial Intelligence

Stichwörter der Autor:innen

Evolutionary roboticscontroller optimizationmorphological evolution

Autor:innen und Institutionen

  • Department of Computer Science, Vrije Universiteit Amsterdam, The Netherlands fuda.van.diggelen@vu.nl.
  • Department of Computer Science, Vrije Universiteit Amsterdam, The Netherlands Technology Innovation Institute, Abu Dhabi, P.O. Box: 9639, Masdar City, UAE e.ferrante@vu.nl.
  • Department of Computer Science, Vrije Universiteit Amsterdam, The Netherlands a.e.eiben@vu.nl.

ORCID-Profile: Fuda van Diggelen, Eliseo Ferrante, A E Eiben

Zitieren

APA

van Diggelen, F., Ferrante, E., & Eiben, A. E. (2024). Comparing Robot Controller Optimization Methods on Evolvable Morphologies. Evolutionary Computation. https://doi.org/10.1162/evco_a_00334

Vancouver

van Diggelen F, Ferrante E, Eiben AE. Comparing Robot Controller Optimization Methods on Evolvable Morphologies. Evolutionary Computation. 2024. doi:10.1162/evco_a_00334

BibTeX
@article{vanDiggelen2024,
  author = {van Diggelen, Fuda and Ferrante, Eliseo and Eiben, A E},
  title = {Comparing Robot Controller Optimization Methods on Evolvable Morphologies},
  journal = {Evolutionary Computation},
  year = {2024},
  doi = {10.1162/evco_a_00334}
}
Kostenlos starten →