Transition cells for navigation and planning in an unknown environment
Nicolas Cuperlier, Mathias Quoy, Christophe Giovanangelli, Philippe Gaussier and Philippe Laroque
Simulation of Adaptive Behavior 2006 (SAB 2006)
Rome, Italy, 25-29 September 2006
Summary
We present a navigation and planning system using vision for extracting non predefined landmarks, a dead-reckoning systems giving the integrated movement and a topological map. Localisation and planning remain possible even if the map is partially unknown. An omnidirectionnal camera gives a panoramic images from which unpredefined landmarks are extracted. The set of landmarks and their azimuths relative to a fixed orientation defines a particular location without any need of an external environment map. Transitions between two locations recognized at time t and t-1 are explicitly coded, and define spatio-temporal transitions. These transitions are the sensory-motor unit chosen to support planning. During exploration, a topological map (our cognitive map) is learned on-line from these transitions without any cartesian coordinates nor occupancy grids. The edges of this map are reinforced when the corresponding path is used and thus enable to take into account dynamical changes of the environment. A very small decay term on these links ensure the coherence of the map. The transitions are linked with the integrated movement used for going from one place to the others. When planning is required, the activities of transitions coding for the required goal in the cognitive map spread along the arcs of this graph giving transitions (vertices) an higher value to the ones closer from this goal. Activities of the map vertices are enough to bias predicted transitions and to obtain the required movement. Behavior of the robot alternates between planning and exploration.