University of Sussex
Browse

File(s) not publicly available

Evolving controllers for a homogeneous system of physical robots: Structured cooperation with minimal sensors

journal contribution
posted on 2023-06-08, 07:07 authored by Matt Quinn, Lincoln Smith, Giles Mayley, Phil HusbandsPhil Husbands
We report on recent work in which we employed artificial evolution to design neural network controllers for small, homogeneous teams of mobile autonomous robots. The robots were evolved to perform a formation-movement task from random starting positions, equipped only with infrared sensors. The dual constraints of homogeneity and minimal sensors make this a non-trivial task. We describe the behaviour of a successful system in which robots adopt and maintain functionally distinct roles in order to achieve the task. We believe this to be the first example of the use of artificial evolution to design coordinated, cooperative behaviour for real robots.

History

Publication status

  • Published

Journal

Philosophical Transactions A: Mathematical, Physical and Engineering Sciences

ISSN

1471-2962

Publisher

Royal Society, The

Issue

1811

Volume

361

Page range

2321-2343

Pages

23.0

Department affiliated with

  • Informatics Publications

Notes

Originality: First successful use of evolutionary robotics methodology to develop coordinated cooperative behaviour in a team of real robots. Rigour: Methodology was extended to allow development of neural controllers for homogeneous teams of autonomous robots equipped with minimal local sensing. Despite severe constraints, the evolved system proved capable of organizing itself as a team, involving functionally distinct roles. Sufficient runs were undertaken to produce statistically highly significant results and prove the efficacy of the methodology. Significance: Demonstrates how to develop non-trivial behaviour with highly constrained minimal systems, with potentially important applications in space science and nano-robotics. Impact: Widely recognised as an important advance in the growing field of evolutionary robotics as evidenced by its inclusion in a number of significant recent international reviews of the field. 29 Google scholar citations, 15 Web of Science. Total Google scholar citations for this journal paper and its preceding conference paper and technical report are 64.

Full text available

  • No

Peer reviewed?

  • Yes

Legacy Posted Date

2012-02-06

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC