SC-CNNs for sensors data fusion and control in space distributed structures

Analog modular architectures, based on the computational paradigm of state-controlled cellular neural networks, are considered to suitably manipulate signals obtained from multiple sensors measurement systems necessary for controlling deformations in space distributed structures. The design methodology to choose the cellular neural network template coefficients to obtain the desired "global" behavior is proposed together with some theoretical results that guarantee asymptotic stability of the system. An experimental prototype of this state-controlled cellular neural network for multisensor data fusion and control applications is presented. Moreover, the problem of controlling the deformation of a multiple link system is tackled.