Finding the Place: How to Train and Use Convolutional Neural Networks for a Dynamically Learning Robot

For a robot, the ability to adapt his knowledge automatically and customize its behavior is a key feature. Furthermore, a robot should be able to carry out its tasks at a long-term basis, performing it seamlessly in presence of changes in their surroundings. To do that, it is essential that the robo...

Volledige beschrijving

Bewaard in:
Bibliografische gegevens
Hoofdauteur: Cruz, Edmanuel (author)
Andere auteurs: Rangel, José Carlos (author), Gomez Donoso, Francisco (author), Bauer, Zuria (author), Cazorla, Miguel (author), García Rodríguez, José (author)
Formaat: article
Taal:Engels
Gepubliceerd in: 2019
Onderwerpen:
Online toegang:https://ieeexplore.ieee.org/abstract/document/8489469/keywords#keywords
https://ridda2.utp.ac.pa/handle/123456789/9438
Tags: Voeg label toe
Geen labels, Wees de eerste die dit record labelt!
_version_ 1869652460794019840
author Cruz, Edmanuel
author2 Rangel, José Carlos
Gomez Donoso, Francisco
Bauer, Zuria
Cazorla, Miguel
García Rodríguez, José
author2_role author
author
author
author
author
author_browse Bauer, Zuria
Cazorla, Miguel
Cruz, Edmanuel
García Rodríguez, José
Gomez Donoso, Francisco
Rangel, José Carlos
author_facet Cruz, Edmanuel
Rangel, José Carlos
Gomez Donoso, Francisco
Bauer, Zuria
Cazorla, Miguel
García Rodríguez, José
author_role author
collection Repositorio Institucional de documento digitales de acceso abierto de la UTP
dc.creator.none.fl_str_mv Cruz, Edmanuel
Rangel, José Carlos
Gomez Donoso, Francisco
Bauer, Zuria
Cazorla, Miguel
García Rodríguez, José
dc.date.none.fl_str_mv 10/15/2018
10/15/2018
2019-12-17T20:50:46Z
2019-12-17T20:50:46Z
2019-12-17T20:50:46Z
2019-12-17T20:50:46Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv https://ieeexplore.ieee.org/abstract/document/8489469/keywords#keywords
2161-4407
https://ridda2.utp.ac.pa/handle/123456789/9438
https://ridda2.utp.ac.pa/handle/123456789/9438
dc.language.none.fl_str_mv eng
en
dc.rights.none.fl_str_mv info:eu-repo/semantics/embargoedAccess
dc.source.none.fl_str_mv reponame:Repositorio Institucional de documento digitales de acceso abierto de la UTP
instname:Universidad Tecnológica de Panamá
instacron:U Tecnológica de Panamá
dc.subject.none.fl_str_mv Robots
Semantics
Training
Feature extraction
Computer architecture
Task analysis
Visualization
Robots
Semantics
Training
Feature extraction
Computer architecture
Task analysis
Visualization
dc.title.none.fl_str_mv Finding the Place: How to Train and Use Convolutional Neural Networks for a Dynamically Learning Robot
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
description For a robot, the ability to adapt his knowledge automatically and customize its behavior is a key feature. Furthermore, a robot should be able to carry out its tasks at a long-term basis, performing it seamlessly in presence of changes in their surroundings. To do that, it is essential that the robot dynamically learn from their environment, but to perform a fully retraining of a deep learning architecture when the model needs new knowledge is a highly time consuming task. This work focus on exploring several strategies to include new data to an already learned model, applied to the semantic localization problem focusing in the accuracy of the final model and their training time. Exhaustive experimentation is carried out and each result is discussed consequently.
eu_rights_str_mv embargoedAccess
format article
id lrtest_3e91ba28f31c51ccda5de614f200a05b
identifier_str_mv 2161-4407
instacron_str U Tecnológica de Panamá
institution U Tecnológica de Panamá
instname_str Universidad Tecnológica de Panamá
language eng
language_invalid_str_mv en
network_acronym_str lrtest
network_name_str lr
oai_identifier_str oai:ridda2.utp.ac.pa:123456789/9438
publishDate 2019
publishDateSort 2019
reponame_str Repositorio Institucional de documento digitales de acceso abierto de la UTP
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Finding the Place: How to Train and Use Convolutional Neural Networks for a Dynamically Learning RobotCruz, EdmanuelRangel, José CarlosGomez Donoso, FranciscoBauer, ZuriaCazorla, MiguelGarcía Rodríguez, JoséRobotsSemanticsTrainingFeature extractionComputer architectureTask analysisVisualizationRobotsSemanticsTrainingFeature extractionComputer architectureTask analysisVisualizationFor a robot, the ability to adapt his knowledge automatically and customize its behavior is a key feature. Furthermore, a robot should be able to carry out its tasks at a long-term basis, performing it seamlessly in presence of changes in their surroundings. To do that, it is essential that the robot dynamically learn from their environment, but to perform a fully retraining of a deep learning architecture when the model needs new knowledge is a highly time consuming task. This work focus on exploring several strategies to include new data to an already learned model, applied to the semantic localization problem focusing in the accuracy of the final model and their training time. Exhaustive experimentation is carried out and each result is discussed consequently.For a robot, the ability to adapt his knowledge automatically and customize its behavior is a key feature. Furthermore, a robot should be able to carry out its tasks at a long-term basis, performing it seamlessly in presence of changes in their surroundings. To do that, it is essential that the robot dynamically learn from their environment, but to perform a fully retraining of a deep learning architecture when the model needs new knowledge is a highly time consuming task. This work focus on exploring several strategies to include new data to an already learned model, applied to the semantic localization problem focusing in the accuracy of the final model and their training time. Exhaustive experimentation is carried out and each result is discussed consequently.2019-12-17T20:50:46Z2019-12-17T20:50:46Z2019-12-17T20:50:46Z2019-12-17T20:50:46Z10/15/201810/15/2018info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://ieeexplore.ieee.org/abstract/document/8489469/keywords#keywords2161-4407https://ridda2.utp.ac.pa/handle/123456789/9438https://ridda2.utp.ac.pa/handle/123456789/9438engeninfo:eu-repo/semantics/embargoedAccessreponame:Repositorio Institucional de documento digitales de acceso abierto de la UTPinstname:Universidad Tecnológica de Panamáinstacron:U Tecnológica de Panamáoai:ridda2.utp.ac.pa:123456789/94382021-07-06T15:35:11Z
spellingShingle Finding the Place: How to Train and Use Convolutional Neural Networks for a Dynamically Learning Robot
Cruz, Edmanuel
Robots
Semantics
Training
Feature extraction
Computer architecture
Task analysis
Visualization
Robots
Semantics
Training
Feature extraction
Computer architecture
Task analysis
Visualization
status_str publishedVersion
title Finding the Place: How to Train and Use Convolutional Neural Networks for a Dynamically Learning Robot
title_full Finding the Place: How to Train and Use Convolutional Neural Networks for a Dynamically Learning Robot
title_fullStr Finding the Place: How to Train and Use Convolutional Neural Networks for a Dynamically Learning Robot
title_full_unstemmed Finding the Place: How to Train and Use Convolutional Neural Networks for a Dynamically Learning Robot
title_short Finding the Place: How to Train and Use Convolutional Neural Networks for a Dynamically Learning Robot
title_sort Finding the Place: How to Train and Use Convolutional Neural Networks for a Dynamically Learning Robot
topic Robots
Semantics
Training
Feature extraction
Computer architecture
Task analysis
Visualization
Robots
Semantics
Training
Feature extraction
Computer architecture
Task analysis
Visualization
url https://ieeexplore.ieee.org/abstract/document/8489469/keywords#keywords
https://ridda2.utp.ac.pa/handle/123456789/9438