Evolving robot's behavior by using CNNs
SAB'06 Proceedings of the 9th international conference on From Animals to Animats: simulation of Adaptive Behavior
Hi-index | 0.00 |
In this paper a reaction-diffusion CNN is implemented togenerate and adaptively control locomotion in a biologically inspiredwalking robot. In particular a dedicated CNN development system hasbeen realised to make the mechatronic device able to select, based onsensory stimuli, the most suitable locomotion type according to theenvironment. The first example of analog implementation of thebiological paradigm of the Central Pattern Generator is thereforepresented.