Evolution of Diverse, Manufacturable Robot Body Plans

Edgar Buchanan; Leni K.Le Goff; Emma Hart; Agoston E. Eiben; Matteo De Carlo; Wei Li; Matthew F. Hale; Mike Angus; Robert Woolley; Alan F. Winfield; Jon Timmis; Andy M. Tyrrell

doi:10.1109/SSCI47803.2020.9308434

Evolution of Diverse, Manufacturable Robot Body Plans

Edgar Buchanan, Leni K.Le Goff, Emma Hart, Agoston E. Eiben, Matteo De Carlo, Wei Li, Matthew F. Hale, Mike Angus, Robert Woolley, Alan F. Winfield, Jon Timmis, Andy M. Tyrrell

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference Proceeding (Non-Journal item)

4 Citations (Scopus)

Abstract

Advances in rapid prototyping have opened up new avenues of research within Evolutionary Robotics in which not only controllers but also the body plans (morphologies) of robots can evolve in real-time and real-space. However, this also introduces new challenges, in that robot models that can be instantiated from an encoding in simulation might not be manufacturable in practice (due to constraints associated with the 3D printing and/or automated assembly processes). We introduce a representation for evolving (wheeled) robots with a printed plastic skeleton, and evaluate three variants of a novelty-search algorithm in terms of their ability to produce populations of manufacturable but diverse robots. While the set of manufacturable robots discovered represent only a small fraction of the overall search space of all robots, all methods are shown to be capable of generating a diverse population of manufacturable robots that we conjecture is large enough to seed an evolving robotic ecosystem.

Original language	English
Title of host publication	2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020
Publisher	IEEE Press
Pages	2132-2139
Number of pages	8
ISBN (Electronic)	9781728125473
DOIs	https://doi.org/10.1109/SSCI47803.2020.9308434
Publication status	Published - 01 Dec 2020
Event	2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020 - Virtual, Canberra, Australia Duration: 01 Dec 2020 → 04 Dec 2020

Publication series

Name	2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020

Conference

Conference	2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020
Country/Territory	Australia
City	Virtual, Canberra
Period	01 Dec 2020 → 04 Dec 2020

Keywords

autonomous robot evolution
autonomous robot fabrication
evolutionary robotics
robot manufacturability

Access to Document

10.1109/SSCI47803.2020.9308434

Permanent link

Persistent link

Cite this

Buchanan, E., Goff, L. K. L., Hart, E., Eiben, A. E., Carlo, M. D., Li, W., Hale, M. F., Angus, M., Woolley, R., Winfield, A. F., Timmis, J., & Tyrrell, A. M. (2020). Evolution of Diverse, Manufacturable Robot Body Plans. In 2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020 (pp. 2132-2139). Article 9308434 (2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020). IEEE Press. https://doi.org/10.1109/SSCI47803.2020.9308434

@inproceedings{149ff1ab94684039b979af7ebefcfbb3,

title = "Evolution of Diverse, Manufacturable Robot Body Plans",

abstract = "Advances in rapid prototyping have opened up new avenues of research within Evolutionary Robotics in which not only controllers but also the body plans (morphologies) of robots can evolve in real-time and real-space. However, this also introduces new challenges, in that robot models that can be instantiated from an encoding in simulation might not be manufacturable in practice (due to constraints associated with the 3D printing and/or automated assembly processes). We introduce a representation for evolving (wheeled) robots with a printed plastic skeleton, and evaluate three variants of a novelty-search algorithm in terms of their ability to produce populations of manufacturable but diverse robots. While the set of manufacturable robots discovered represent only a small fraction of the overall search space of all robots, all methods are shown to be capable of generating a diverse population of manufacturable robots that we conjecture is large enough to seed an evolving robotic ecosystem.",

keywords = "autonomous robot evolution, autonomous robot fabrication, evolutionary robotics, robot manufacturability",

author = "Edgar Buchanan and Goff, {Leni K.Le} and Emma Hart and Eiben, {Agoston E.} and Carlo, {Matteo De} and Wei Li and Hale, {Matthew F.} and Mike Angus and Robert Woolley and Winfield, {Alan F.} and Jon Timmis and Tyrrell, {Andy M.}",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020 ; Conference date: 01-12-2020 Through 04-12-2020",

year = "2020",

month = dec,

day = "1",

doi = "10.1109/SSCI47803.2020.9308434",

language = "English",

series = "2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020",

publisher = "IEEE Press",

pages = "2132--2139",

booktitle = "2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020",

address = "United States of America",

}

Buchanan, E, Goff, LKL, Hart, E, Eiben, AE, Carlo, MD, Li, W, Hale, MF, Angus, M, Woolley, R, Winfield, AF, Timmis, J & Tyrrell, AM 2020, Evolution of Diverse, Manufacturable Robot Body Plans. in 2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020., 9308434, 2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020, IEEE Press, pp. 2132-2139, 2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020, Virtual, Canberra, Australia, 01 Dec 2020. https://doi.org/10.1109/SSCI47803.2020.9308434

Evolution of Diverse, Manufacturable Robot Body Plans. / Buchanan, Edgar; Goff, Leni K.Le; Hart, Emma et al.
2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020. IEEE Press, 2020. p. 2132-2139 9308434 (2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020).

Research output: Chapter in Book/Report/Conference proceeding › Conference Proceeding (Non-Journal item)

TY - GEN

T1 - Evolution of Diverse, Manufacturable Robot Body Plans

AU - Buchanan, Edgar

AU - Goff, Leni K.Le

AU - Hart, Emma

AU - Eiben, Agoston E.

AU - Carlo, Matteo De

AU - Li, Wei

AU - Hale, Matthew F.

AU - Angus, Mike

AU - Woolley, Robert

AU - Winfield, Alan F.

AU - Timmis, Jon

AU - Tyrrell, Andy M.

PY - 2020/12/1

Y1 - 2020/12/1

N2 - Advances in rapid prototyping have opened up new avenues of research within Evolutionary Robotics in which not only controllers but also the body plans (morphologies) of robots can evolve in real-time and real-space. However, this also introduces new challenges, in that robot models that can be instantiated from an encoding in simulation might not be manufacturable in practice (due to constraints associated with the 3D printing and/or automated assembly processes). We introduce a representation for evolving (wheeled) robots with a printed plastic skeleton, and evaluate three variants of a novelty-search algorithm in terms of their ability to produce populations of manufacturable but diverse robots. While the set of manufacturable robots discovered represent only a small fraction of the overall search space of all robots, all methods are shown to be capable of generating a diverse population of manufacturable robots that we conjecture is large enough to seed an evolving robotic ecosystem.

AB - Advances in rapid prototyping have opened up new avenues of research within Evolutionary Robotics in which not only controllers but also the body plans (morphologies) of robots can evolve in real-time and real-space. However, this also introduces new challenges, in that robot models that can be instantiated from an encoding in simulation might not be manufacturable in practice (due to constraints associated with the 3D printing and/or automated assembly processes). We introduce a representation for evolving (wheeled) robots with a printed plastic skeleton, and evaluate three variants of a novelty-search algorithm in terms of their ability to produce populations of manufacturable but diverse robots. While the set of manufacturable robots discovered represent only a small fraction of the overall search space of all robots, all methods are shown to be capable of generating a diverse population of manufacturable robots that we conjecture is large enough to seed an evolving robotic ecosystem.

KW - autonomous robot evolution

KW - autonomous robot fabrication

KW - evolutionary robotics

KW - robot manufacturability

UR - http://www.scopus.com/inward/record.url?scp=85097817004&partnerID=8YFLogxK

U2 - 10.1109/SSCI47803.2020.9308434

DO - 10.1109/SSCI47803.2020.9308434

M3 - Conference Proceeding (Non-Journal item)

AN - SCOPUS:85097817004

T3 - 2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020

SP - 2132

EP - 2139

BT - 2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020

PB - IEEE Press

T2 - 2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020

Y2 - 1 December 2020 through 4 December 2020

ER -

Evolution of Diverse, Manufacturable Robot Body Plans

Abstract

Publication series

Conference

Keywords

Access to Document

Permanent link

Other files and links

Fingerprint

Cite this