A mobile robot that learns its place
Neural Computation
Learning metric-topological maps for indoor mobile robot navigation
Artificial Intelligence
Learning topological maps with weak local odometric information
IJCAI'97 Proceedings of the Fifteenth international joint conference on Artifical intelligence - Volume 2
2005 Special issue: Robust self-localisation and navigation based on hippocampal place cells
Neural Networks - Special issue: Computational theories of the functions of the hippocampus
Stability of localized patterns in neural fields
Neural Computation
A neurobiologically motivated model for self-organized learning
MICAI'05 Proceedings of the 4th Mexican international conference on Advances in Artificial Intelligence
Hi-index | 0.00 |
We present a framework for constructing representations ofspace in an autonomous agent which does not obtain any directinformation about its location. Instead the algorithm reliesexclusively on inputs from its sensors. Activations within a neuralnetwork are propagated in time depending on the input from receptorswhich signal the agent‘s own actions. The connections of the networkto receptors for external stimuli are adapted according to a Hebbianlearning rule derived from the prediction error on sensory inputsone time step ahead. During exploration of the environment therespective cells become selectively activated by particular locationsand directions even when relying on highly ambiguous stimuli.