Predicting the internal model of a robotic system from its morphology
![Thumbnail](/xmlui/bitstream/handle/10234/180786/61839_postprint.pdf.jpg?sequence=7&isAllowed=y)
View/ Open
Metadata
Show full item recordcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/7036
comunitat-uji-handle3:10234/8620
comunitat-uji-handle4:
INVESTIGACIONMetadata
Title
Predicting the internal model of a robotic system from its morphologyDate
2018Publisher
ElsevierISSN
0921-8890Bibliographic citation
DURAN, Angel J.; DEL POBIL, Angel P. Predicting the internal model of a robotic system from its morphology. Robotics and Autonomous Systems, 2018, vol. 110: 33-43Type
info:eu-repo/semantics/articlePublisher version
https://www.sciencedirect.com/science/article/pii/S0921889017306942Version
info:eu-repo/semantics/submittedVersionSubject
Abstract
The estimation of the internal model of a robotic system results from the interaction of its morphology, sensors and
actuators, with a particular environment. Model learning techniques, based on supervised machine ... [+]
The estimation of the internal model of a robotic system results from the interaction of its morphology, sensors and
actuators, with a particular environment. Model learning techniques, based on supervised machine learning, are
widespread for determining the internal model. An important limitation of such approaches is that once a model has
been learnt, it does not behave properly when the robot morphology is changed. From this it follows that there must
exist a relationship between them. We propose a model for this correlation between the morphology and the internal
model parameters, so that a new internal model can be predicted when the morphological parameters are modified.
Di erent neural network architectures are proposed to address this high dimensional regression problem. A case
study is analyzed in detail to illustrate and evaluate the performance of the approach, namely, a pan-tilt robot head
executing saccadic movements. The best results are obtained for an architecture with parallel neural networks due
to the independence of its outputs. Theses results can have a great significance since the predicted parameters can
dramatically speed up the adaptation process following a change in morphology [-]
Is part of
Robotics and Autonomous Systems, 2018, vol. 110: 33-43Investigation project
This paper describes research done at the UJI Robotic Intelligence Laboratory. Support for this laboratory is provided in part by Ministerio de Economa y Competitividad (DPI2015-69041-R), by Fondo Europeo de Desarrollo Regional (FEDER), by Generalitat Valenciana (PROMETEOII/2014/028) and by Universitat Jaume I (P1-1B2014-52, PREDOC/2013/06).Rights
info:eu-repo/semantics/openAccess
This item appears in the folowing collection(s)
- ICC_Articles [424]