A Muti-Functional Dynamic Neural Processor for Control Applications

In this paper we propose a neural network structure called dynamic neural processor (DNP) which comprises of two dynamic neural units coupled as excitatory and inhibitory neurons. This neural model is inspired by the collective computation of subpopulation of biological neurons. It is demonstrated in this paper that the proposed neural architecture can perform various functions, such as learning the inverse kinematics transformation of two- and three-linked robots, and controlling the unknown nonlinear dynamic systems.