Publicação

Evolving Self-Assembly in Autonomous Homogeneous Robots: Experiments with Two Physical Robots

Ver documento

Detalhes bibliográficos
Resumo:This research work illustrates an approach to the design of controllers for self-assembling robots in which the self-assembly is initiated and regulated by perceptual cues that are brought forth by the physical robots through their dynamical interactions. More specifically, we present a homogeneous control system that can achieve assembly between two modules (two fully autonomous robots) of a mobile self-reconfigurable system without a priori introduced behavioral or morphological heterogeneities. The controllers are dynamic neural networks evolved in simulation that directly control all the actuators of the two robots. The neurocontrollers cause the dynamic specialization of the robots by allocating roles between them based solely on their interaction. We show that the best evolved controller proves to be successful when tested on a real hardware platform, the swarm-bot. The performance achieved is similar to the one achieved by existing modular or behavior-based approaches, also due to the effect of an emergent recovery mechanism that was neither explicitly rewarded by the fitness function, nor observed during the evolutionary simulation. Our results suggest that direct access to the orientations or intentions of the other agents is not a necessary condition for robot coordination: Our robots coordinate without direct or explicit communication, contrary to what is assumed by most research works in collective robotics. This work also contributes to strengthening the evidence that evolutionary robotics is a design methodology that can tackle real-world tasks demanding fine sensory-motor coordination.
Autores principais:Ampatzis, Christos
Outros Autores:Tuci, Elio; Trianni, Vito; Christensen, Anders Lyhne; Dorigo, Marco
Assunto:Self-assembly Role allocation Neural network Artificial evolution Evolutionary robotics
Ano:2009
País:Portugal
Tipo de documento:artigo
Tipo de acesso:acesso aberto
Instituição associada:ISCTE
Idioma:inglês
Origem:Repositório ISCTE
_version_ 1868443548427223040
author Ampatzis, Christos
author2 Tuci, Elio
Trianni, Vito
Christensen, Anders Lyhne
Dorigo, Marco
author2_role author
author
author
author
author_facet Ampatzis, Christos
Tuci, Elio
Trianni, Vito
Christensen, Anders Lyhne
Dorigo, Marco
author_role author
country_str PT
creators_json_txt [{\"Person.name\":\"Ampatzis, Christos\"},{\"Person.name\":\"Tuci, Elio\"},{\"Person.name\":\"Trianni, Vito\"},{\"Person.name\":\"Christensen, Anders Lyhne\"},{\"Person.name\":\"Dorigo, Marco\"}]
datacite.creators.creator.creatorName.fl_str_mv Ampatzis, Christos
Tuci, Elio
Trianni, Vito
Christensen, Anders Lyhne
Dorigo, Marco
datacite.date.Accepted.fl_str_mv 2009-07-24T00:00:00Z
datacite.date.available.fl_str_mv 2013-08-12T15:06:57Z
datacite.date.embargoed.fl_str_mv 2013-08-12T15:06:57Z
datacite.rights.fl_str_mv http://purl.org/coar/access_right/c_abf2
datacite.subjects.subject.fl_str_mv Self-assembly
Role allocation
Neural network
Artificial evolution
Evolutionary robotics
datacite.titles.title.fl_str_mv Evolving Self-Assembly in Autonomous Homogeneous Robots: Experiments with Two Physical Robots
dc.creator.none.fl_str_mv Ampatzis, Christos
Tuci, Elio
Trianni, Vito
Christensen, Anders Lyhne
Dorigo, Marco
dc.date.Accepted.fl_str_mv 2009-07-24T00:00:00Z
dc.date.available.fl_str_mv 2013-08-12T15:06:57Z
dc.date.embargoed.fl_str_mv 2013-08-12T15:06:57Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv http://hdl.handle.net/10071/5439
dc.language.none.fl_str_mv eng
dc.publisher.none.fl_str_mv Massachusetts Institute of Technology
dc.rights.none.fl_str_mv http://purl.org/coar/access_right/c_abf2
dc.subject.none.fl_str_mv Self-assembly
Role allocation
Neural network
Artificial evolution
Evolutionary robotics
dc.title.fl_str_mv Evolving Self-Assembly in Autonomous Homogeneous Robots: Experiments with Two Physical Robots
dc.type.none.fl_str_mv http://purl.org/coar/resource_type/c_6501
description This research work illustrates an approach to the design of controllers for self-assembling robots in which the self-assembly is initiated and regulated by perceptual cues that are brought forth by the physical robots through their dynamical interactions. More specifically, we present a homogeneous control system that can achieve assembly between two modules (two fully autonomous robots) of a mobile self-reconfigurable system without a priori introduced behavioral or morphological heterogeneities. The controllers are dynamic neural networks evolved in simulation that directly control all the actuators of the two robots. The neurocontrollers cause the dynamic specialization of the robots by allocating roles between them based solely on their interaction. We show that the best evolved controller proves to be successful when tested on a real hardware platform, the swarm-bot. The performance achieved is similar to the one achieved by existing modular or behavior-based approaches, also due to the effect of an emergent recovery mechanism that was neither explicitly rewarded by the fitness function, nor observed during the evolutionary simulation. Our results suggest that direct access to the orientations or intentions of the other agents is not a necessary condition for robot coordination: Our robots coordinate without direct or explicit communication, contrary to what is assumed by most research works in collective robotics. This work also contributes to strengthening the evidence that evolutionary robotics is a design methodology that can tackle real-world tasks demanding fine sensory-motor coordination.
dirty 0
eu_rights_str_mv openAccess
format article
id iscte_4a2be1dfd504776fbac6ab4e8953732a
identifier.url.fl_str_mv http://hdl.handle.net/10071/5439
instacron_str iscte
institution ISCTE
instname_str ISCTE
language eng
network_acronym_str iscte
network_name_str Repositório ISCTE
oai_identifier_str oai:repositorio.iscte-iul.pt:10071/5439
organization_str_mv urn:organizationAcronym:iscte
person_str_mv Ampatzis, Christos
Tuci, Elio
Trianni, Vito
Christensen, Anders Lyhne
Dorigo, Marco
publishDate 2009
publisher.none.fl_str_mv Massachusetts Institute of Technology
reponame_str Repositório ISCTE
repository_id_str urn:repositoryAcronym:iscte
service_str_mv urn:repositoryAcronym:iscte
spelling porThis research work illustrates an approach to the design of controllers for self-assembling robots in which the self-assembly is initiated and regulated by perceptual cues that are brought forth by the physical robots through their dynamical interactions. More specifically, we present a homogeneous control system that can achieve assembly between two modules (two fully autonomous robots) of a mobile self-reconfigurable system without a priori introduced behavioral or morphological heterogeneities. The controllers are dynamic neural networks evolved in simulation that directly control all the actuators of the two robots. The neurocontrollers cause the dynamic specialization of the robots by allocating roles between them based solely on their interaction. We show that the best evolved controller proves to be successful when tested on a real hardware platform, the swarm-bot. The performance achieved is similar to the one achieved by existing modular or behavior-based approaches, also due to the effect of an emergent recovery mechanism that was neither explicitly rewarded by the fitness function, nor observed during the evolutionary simulation. Our results suggest that direct access to the orientations or intentions of the other agents is not a necessary condition for robot coordination: Our robots coordinate without direct or explicit communication, contrary to what is assumed by most research works in collective robotics. This work also contributes to strengthening the evidence that evolutionary robotics is a design methodology that can tackle real-world tasks demanding fine sensory-motor coordination.application/pdfengMassachusetts Institute of TechnologyporEvolving Self-Assembly in Autonomous Homogeneous Robots: Experiments with Two Physical RobotsAmpatzis, ChristosTuci, ElioTrianni, VitoChristensen, Anders LyhneDorigo, MarcoHandlehttp://hdl.handle.net/10071/5439ISSNIsPartOf1064-54622013-08-12T15:06:57Z2009-07-24T00:00:00Z2009-07-24http://purl.org/coar/access_right/c_abf2open accessporSelf-assemblyporRole allocationporNeural networkporArtificial evolutionporEvolutionary robotics364445 byteshttp://purl.org/coar/access_right/c_abf2application/pdffulltexthttps://repositorio.iscte-iul.pt/bitstreams/6627bb99-31d0-4ee1-9504-ebbe1fe991ce/downloadliteraturehttp://purl.org/coar/resource_type/c_6501journal article
spellingShingle Evolving Self-Assembly in Autonomous Homogeneous Robots: Experiments with Two Physical Robots
Ampatzis, Christos
Self-assembly
Role allocation
Neural network
Artificial evolution
Evolutionary robotics
status SINGLETON
subject.fl_str_mv Self-assembly
Role allocation
Neural network
Artificial evolution
Evolutionary robotics
title Evolving Self-Assembly in Autonomous Homogeneous Robots: Experiments with Two Physical Robots
title_full Evolving Self-Assembly in Autonomous Homogeneous Robots: Experiments with Two Physical Robots
title_fullStr Evolving Self-Assembly in Autonomous Homogeneous Robots: Experiments with Two Physical Robots
title_full_unstemmed Evolving Self-Assembly in Autonomous Homogeneous Robots: Experiments with Two Physical Robots
title_short Evolving Self-Assembly in Autonomous Homogeneous Robots: Experiments with Two Physical Robots
title_sort Evolving Self-Assembly in Autonomous Homogeneous Robots: Experiments with Two Physical Robots
topic Self-assembly
Role allocation
Neural network
Artificial evolution
Evolutionary robotics
topic_facet Self-assembly
Role allocation
Neural network
Artificial evolution
Evolutionary robotics
url http://hdl.handle.net/10071/5439
visible 1