Skip to Main content Skip to Navigation
Theses

Mécanismes d'inspiration corticale pour l'apprentissage et la représentation d'asservissements sensori-moteurs en robotique

Olivier Ménard 1
1 CORTEX - Neuromimetic intelligence
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : This thesis aims at setting up a sensory-motor loop, that can be used
in robotics, while using biologically, and more precisely cortically,
inspired algorithms. We hope this method allows to reproduce the
robustness, structural uniformity and adaptability that are some of
the most remarkable qualities of the cortical substrate. From a
computational point of view, our model is based on extensions of the
cellular automata. This brings us to program units within maps, each
of these maps representing a biological cortical map. We design
updating mechanisms for these units leading to emergent effects within
the maps, so that sensory-motor loops stand.

In addition, we think that perceiving is the same as preparing an
action. We therefore have to balance our holistic view of perception
with the fact that the multiple modalities of perception are each
represented on a different map in our model, as it is the case in the
cortex. That is why the major part of this thesis aims at creating an
algorithm designed to link the modal maps, while keeping the number of
links low for computational reasons. This creates strong constraints
on both the organization and the learning algorithm of our modal
maps. While we used strictly local computations, each unit, in each
map in our model, realizes a compromise between local influences of
the map and influences from the other maps, that allow our model to
keep a multi-modal coherence.
Document type :
Theses
Complete list of metadatas

Cited literature [124 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00118053
Contributor : Frédéric Alexandre <>
Submitted on : Monday, December 4, 2006 - 10:42:09 AM
Last modification on : Monday, October 19, 2020 - 11:12:58 AM
Long-term archiving on: : Tuesday, April 6, 2010 - 7:21:22 PM

Identifiers

  • HAL Id : tel-00118053, version 1

Collections

Citation

Olivier Ménard. Mécanismes d'inspiration corticale pour l'apprentissage et la représentation d'asservissements sensori-moteurs en robotique. Interface homme-machine [cs.HC]. Université Henri Poincaré - Nancy I, 2006. Français. ⟨tel-00118053⟩

Share

Metrics

Record views

278

Files downloads

709