Name
Abstract | ||
|---|---|---|
In this article, we apply a bio-inspired control architecture toa roving robot performing different tasks. The key of the controlsystem is the perceptual core, where heterogeneous informationcoming from sensors is merged to build an internal portraitrepresenting the current situation of the environment. The internalrepresentation triggers an action as the response to the currentstimuli, closing the loop between the agent and the external world.The robot's internal state is implemented through a nonlinearlattice of neuron cells, allowing the generation of a large amountof emergent steady-state solutions in the form of Turing patterns.These are incrementally shaped, through learning, so as toconstitute a "mirror" of the environmental conditions.Reaction-diffusion cellular nonlinear networks were chosen togenerate Turing patterns as internal representations of the robotsurroundings. The associations between incoming sensations and theperceptual core, and between Turing patterns and actions to beperformed, are driven by two reward-based learning mechanisms. Wereport on simulation results and experiments on a roving robot toshow the suitability of the approach. |
Year | DOI | Venue |
|---|---|---|
2008 | 10.1177/1059712308089181 | Adaptive Behaviour |
Keywords | Field | DocType |
dynamic spatiotemporal patterns applied,perceptual core,internal representation,roving robot toshow,bio-inspired control architecture toa,theperceptual core,turing pattern,reaction-diffusion cellular nonlinear network,internal state,roving robot,togenerate turing pattern,reward-based learning mechanism,control system,steady state,perception,reaction diffusion | Nonlinear system,Computer science,Turing patterns,Nonlinear dynamical systems,Artificial intelligence,Stimulus (physiology),Control system,Robot,Perception,Machine learning,Nonlinear networks | Journal |
Volume | Issue | ISSN |
16 | 2-3 | 1059-7123 |
Citations | PageRank | References |
6 | 0.58 | 11 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Paolo Arena | 1 | 261 | 47.43 |
| Luigi Fortuna | 2 | 761 | 128.37 |
| Davide Lombardo | 3 | 17 | 2.16 |
| Luca Patané | 4 | 104 | 17.31 |