Comparing encodings for performance and phenotypic exploration in evolving modular robots

Frank Veenstra, Emma Hart, Edgar Buchanan, Wei Li, Matteo De Carlo, Ágoston E. EibenGECCO (Companion) 2019 · OpenReview

Abstract

To investigate how encodings influence evolving the morphology and control of modular robots, we compared three encodings: a direct encoding and two generative encodings---a compositional pattern producing network (CPPN) and a Lindenmayer System (L-System). The evolutionary progression and final performance of the direct encoding and the L-System was significantly better than the CPPN. The generative encodings converge quicker than the direct encoding in terms of morphological and controller diversity.
BibTeX
@inproceedings{veenstra2019comparing,
  author = {Frank Veenstra and Emma Hart and Edgar Buchanan and Wei Li and Matteo De Carlo and Ágoston E. Eiben},
  title = {{Comparing encodings for performance and phenotypic exploration in evolving modular robots}},
  booktitle = {GECCO (Companion)},
  year = {2019},
  url = {https://openreview.net/forum?id=1eltKjq4hho},
}

Loading related content...

Figure