Navigation on Topologic Feature-Maps
IIZUKA ´94, Fukuoka, Japan August 1-7, 1994
Uwe R. Zimmer, Cornelia Fischer & Ewald von Puttkamer - contact authors
Based on the idea of using topologic feature-maps instead of geometric
environment maps in practical mobile robot tasks, we show an applicable
way to
navigate on such topologic maps. The main features regarding this kind of
navigation are: handling of very inaccurate position (and orientation)
information as well as implicit modelling of complex kinematics during an
adaptation phase.
Due to the lack of proper a-priori knowledge, a reinforcement based model
is
used for the translation of navigator commands to motor actions. Instead
of
employing a backpropagation network for the central associative memory
module
(attaching action-probabilities to sensor situations resp. navigator
commands) a
much faster dynamic cell structure system based on dynamic feature maps
is
shown. Standard graph-search heuristics like A* are applied in the
planning
phase.