Evolution of swarm robotics systems with novelty search
Main Author: | |
---|---|
Publication Date: | 2013 |
Other Authors: | , |
Format: | Article |
Language: | eng |
Source: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Download full: | https://ciencia.iscte-iul.pt/id/ci-pub-13595 http://hdl.handle.net/10071/14017 |
Summary: | Novelty search is a recent artificial evolution technique that challenges traditional evolutionary approaches. In novelty search, solutions are rewarded based on their novelty, rather than their quality with respect to a predefined objective. The lack of a predefined objective precludes premature convergence caused by a deceptive fitness function. In this paper, we apply novelty search combined with NEAT to the evolution of neural controllers for homogeneous swarms of robots. Our empirical study is conducted in simulation, and we use a common swarm robotics task—aggregation, and a more challenging task—sharing of an energy recharging station. Our results show that novelty search is unaffected by deception, is notably effective in bootstrapping evolution, can find solutions with lower complexity than fitness-based evolution, and can find a broad diversity of solutions for the same task. Even in non-deceptive setups, novelty search achieves solution qualities similar to those obtained in traditional fitness-based evolution. Our study also encompasses variants of novelty search that work in concert with fitness-based evolution to combine the exploratory character of novelty search with the exploitatory character of objective-based evolution. We show that these variants can further improve the performance of novelty search. Overall, our study shows that novelty search is a promising alternative for the evolution of controllers for robotic swarms. |
id |
RCAP_1ab23d0ebc7e7a11ed187669b8e98508 |
---|---|
oai_identifier_str |
oai:repositorio.iscte-iul.pt:10071/14017 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Evolution of swarm robotics systems with novelty searchEvolutionary roboticsNeuroevolutionSwarm roboticsNovelty searchNEATBehavioural diversityDeceptionNovelty search is a recent artificial evolution technique that challenges traditional evolutionary approaches. In novelty search, solutions are rewarded based on their novelty, rather than their quality with respect to a predefined objective. The lack of a predefined objective precludes premature convergence caused by a deceptive fitness function. In this paper, we apply novelty search combined with NEAT to the evolution of neural controllers for homogeneous swarms of robots. Our empirical study is conducted in simulation, and we use a common swarm robotics task—aggregation, and a more challenging task—sharing of an energy recharging station. Our results show that novelty search is unaffected by deception, is notably effective in bootstrapping evolution, can find solutions with lower complexity than fitness-based evolution, and can find a broad diversity of solutions for the same task. Even in non-deceptive setups, novelty search achieves solution qualities similar to those obtained in traditional fitness-based evolution. Our study also encompasses variants of novelty search that work in concert with fitness-based evolution to combine the exploratory character of novelty search with the exploitatory character of objective-based evolution. We show that these variants can further improve the performance of novelty search. Overall, our study shows that novelty search is a promising alternative for the evolution of controllers for robotic swarms.Springer US2017-07-14T09:32:54Z2013-01-01T00:00:00Z20132017-07-14T09:31:14Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfhttps://ciencia.iscte-iul.pt/id/ci-pub-13595http://hdl.handle.net/10071/14017eng1935-381210.1007/s11721-013-0081-zGomes, J.Urbano, P.Christensen, A. L.info:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-11-09T17:59:06Zoai:repositorio.iscte-iul.pt:10071/14017Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T22:30:56.922078Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
Evolution of swarm robotics systems with novelty search |
title |
Evolution of swarm robotics systems with novelty search |
spellingShingle |
Evolution of swarm robotics systems with novelty search Gomes, J. Evolutionary robotics Neuroevolution Swarm robotics Novelty search NEAT Behavioural diversity Deception |
title_short |
Evolution of swarm robotics systems with novelty search |
title_full |
Evolution of swarm robotics systems with novelty search |
title_fullStr |
Evolution of swarm robotics systems with novelty search |
title_full_unstemmed |
Evolution of swarm robotics systems with novelty search |
title_sort |
Evolution of swarm robotics systems with novelty search |
author |
Gomes, J. |
author_facet |
Gomes, J. Urbano, P. Christensen, A. L. |
author_role |
author |
author2 |
Urbano, P. Christensen, A. L. |
author2_role |
author author |
dc.contributor.author.fl_str_mv |
Gomes, J. Urbano, P. Christensen, A. L. |
dc.subject.por.fl_str_mv |
Evolutionary robotics Neuroevolution Swarm robotics Novelty search NEAT Behavioural diversity Deception |
topic |
Evolutionary robotics Neuroevolution Swarm robotics Novelty search NEAT Behavioural diversity Deception |
description |
Novelty search is a recent artificial evolution technique that challenges traditional evolutionary approaches. In novelty search, solutions are rewarded based on their novelty, rather than their quality with respect to a predefined objective. The lack of a predefined objective precludes premature convergence caused by a deceptive fitness function. In this paper, we apply novelty search combined with NEAT to the evolution of neural controllers for homogeneous swarms of robots. Our empirical study is conducted in simulation, and we use a common swarm robotics task—aggregation, and a more challenging task—sharing of an energy recharging station. Our results show that novelty search is unaffected by deception, is notably effective in bootstrapping evolution, can find solutions with lower complexity than fitness-based evolution, and can find a broad diversity of solutions for the same task. Even in non-deceptive setups, novelty search achieves solution qualities similar to those obtained in traditional fitness-based evolution. Our study also encompasses variants of novelty search that work in concert with fitness-based evolution to combine the exploratory character of novelty search with the exploitatory character of objective-based evolution. We show that these variants can further improve the performance of novelty search. Overall, our study shows that novelty search is a promising alternative for the evolution of controllers for robotic swarms. |
publishDate |
2013 |
dc.date.none.fl_str_mv |
2013-01-01T00:00:00Z 2013 2017-07-14T09:32:54Z 2017-07-14T09:31:14Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://ciencia.iscte-iul.pt/id/ci-pub-13595 http://hdl.handle.net/10071/14017 |
url |
https://ciencia.iscte-iul.pt/id/ci-pub-13595 http://hdl.handle.net/10071/14017 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
1935-3812 10.1007/s11721-013-0081-z |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Springer US |
publisher.none.fl_str_mv |
Springer US |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799134871327604736 |