Header logo is

View-based cognitive map learning by an autonomous robot

1995

Conference Paper

ei


This paper presents a view-based approach to map learning and navigation in mazes. By means of graph theory we have shown that the view-graph is a sufficient representation for map behaviour such as path planning. A neural network for unsupervised learning of the view-graph from sequences of views is constructed. We use a modified Kohonen (1988) learning rule that transforms temporal sequence (rather than featural similarity) into connectedness. In the main part of the paper, we present a robot implementation of the scheme. The results show that the proposed network is able to support map behaviour in simple environments.

Author(s): Mallot, HA. and Bülthoff, HH. and Georg, P. and Schölkopf, B. and Yasuhara, K.
Book Title: Proceedings International Conference on Artificial Neural Networks, vol. 2
Journal: Proceedings International Conference on Artificial Neural Networks (ICANN
Pages: 381-386
Year: 1995
Month: October
Day: 0
Editors: Fogelman-Soulié, F.
Publisher: EC2

Department(s): Empirische Inferenz
Bibtex Type: Conference Paper (inproceedings)

Event Name: Conférence Internationale sur les Réseaux de Neurones Artificiels (ICANN ’95)
Event Place: Paris, France

Address: Paris, France
Digital: 0
ISBN: 2-910085-18-X
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF

BibTex

@inproceedings{668,
  title = {View-based cognitive map learning by an autonomous robot},
  author = {Mallot, HA. and B{\"u}lthoff, HH. and Georg, P. and Sch{\"o}lkopf, B. and Yasuhara, K.},
  journal = {Proceedings International Conference on Artificial Neural Networks (ICANN},
  booktitle = {Proceedings International Conference on Artificial Neural Networks, vol. 2},
  pages = {381-386},
  editors = {Fogelman-Soulié, F.},
  publisher = {EC2},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  address = {Paris, France},
  month = oct,
  year = {1995},
  doi = {},
  month_numeric = {10}
}