Cargando…

Simulated mental imagery for robotic task planning

Traditional AI-planning methods for task planning in robotics require a symbolically encoded domain description. While powerful in well-defined scenarios, as well as human-interpretable, setting this up requires a substantial effort. Different from this, most everyday planning tasks are solved by hu...

Descripción completa

Detalles Bibliográficos
Autores principales: Li, Shijia, Kulvicius, Tomas, Tamosiunaite, Minija, Wörgötter, Florentin
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10483834/
https://www.ncbi.nlm.nih.gov/pubmed/37692886
http://dx.doi.org/10.3389/fnbot.2023.1218977
_version_ 1785102469944049664
author Li, Shijia
Kulvicius, Tomas
Tamosiunaite, Minija
Wörgötter, Florentin
author_facet Li, Shijia
Kulvicius, Tomas
Tamosiunaite, Minija
Wörgötter, Florentin
author_sort Li, Shijia
collection PubMed
description Traditional AI-planning methods for task planning in robotics require a symbolically encoded domain description. While powerful in well-defined scenarios, as well as human-interpretable, setting this up requires a substantial effort. Different from this, most everyday planning tasks are solved by humans intuitively, using mental imagery of the different planning steps. Here, we suggest that the same approach can be used for robots too, in cases which require only limited execution accuracy. In the current study, we propose a novel sub-symbolic method called Simulated Mental Imagery for Planning (SiMIP), which consists of perception, simulated action, success checking, and re-planning performed on 'imagined' images. We show that it is possible to implement mental imagery-based planning in an algorithmically sound way by combining regular convolutional neural networks and generative adversarial networks. With this method, the robot acquires the capability to use the initially existing scene to generate action plans without symbolic domain descriptions, while at the same time, plans remain human-interpretable, different from deep reinforcement learning, which is an alternative sub-symbolic approach. We create a data set from real scenes for a packing problem of having to correctly place different objects into different target slots. This way efficiency and success rate of this algorithm could be quantified.
format Online
Article
Text
id pubmed-10483834
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-104838342023-09-08 Simulated mental imagery for robotic task planning Li, Shijia Kulvicius, Tomas Tamosiunaite, Minija Wörgötter, Florentin Front Neurorobot Neuroscience Traditional AI-planning methods for task planning in robotics require a symbolically encoded domain description. While powerful in well-defined scenarios, as well as human-interpretable, setting this up requires a substantial effort. Different from this, most everyday planning tasks are solved by humans intuitively, using mental imagery of the different planning steps. Here, we suggest that the same approach can be used for robots too, in cases which require only limited execution accuracy. In the current study, we propose a novel sub-symbolic method called Simulated Mental Imagery for Planning (SiMIP), which consists of perception, simulated action, success checking, and re-planning performed on 'imagined' images. We show that it is possible to implement mental imagery-based planning in an algorithmically sound way by combining regular convolutional neural networks and generative adversarial networks. With this method, the robot acquires the capability to use the initially existing scene to generate action plans without symbolic domain descriptions, while at the same time, plans remain human-interpretable, different from deep reinforcement learning, which is an alternative sub-symbolic approach. We create a data set from real scenes for a packing problem of having to correctly place different objects into different target slots. This way efficiency and success rate of this algorithm could be quantified. Frontiers Media S.A. 2023-08-24 /pmc/articles/PMC10483834/ /pubmed/37692886 http://dx.doi.org/10.3389/fnbot.2023.1218977 Text en Copyright © 2023 Li, Kulvicius, Tamosiunaite and Wörgötter. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Neuroscience
Li, Shijia
Kulvicius, Tomas
Tamosiunaite, Minija
Wörgötter, Florentin
Simulated mental imagery for robotic task planning
title Simulated mental imagery for robotic task planning
title_full Simulated mental imagery for robotic task planning
title_fullStr Simulated mental imagery for robotic task planning
title_full_unstemmed Simulated mental imagery for robotic task planning
title_short Simulated mental imagery for robotic task planning
title_sort simulated mental imagery for robotic task planning
topic Neuroscience
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10483834/
https://www.ncbi.nlm.nih.gov/pubmed/37692886
http://dx.doi.org/10.3389/fnbot.2023.1218977
work_keys_str_mv AT lishijia simulatedmentalimageryforrobotictaskplanning
AT kulviciustomas simulatedmentalimageryforrobotictaskplanning
AT tamosiunaiteminija simulatedmentalimageryforrobotictaskplanning
AT worgotterflorentin simulatedmentalimageryforrobotictaskplanning