Collective behaviour in cellular neural networks to model the central pattern generator

Abstract

In this paper, the synergetic aspects of locally interacting nonlinear systems, called cellular neural networks (CNNs), are considered. In particular, the reaction-diffusion dynamics are analysed in order to generate self-organizing phenomena encountered in nature, such as autonomous waves as well as morphogenetical patterns. These aspects are finally considered to model, in a very schematic and simple way, the role of the central pattern generator (CPG) for the generation and control of locomotion in animals. Following these guidelines, the CPG is modelled using CNNs. Some applications are presented to the locomotion coordination of some walking robot prototypes. The approach proposed is very efficient since it allows us to obtain complex artificial locomotion control strategies on chip.