A neural processor for coordinating multiple systems with dynamic uncertainties

The authors describe a neural model, called the dynamic neural processor (DNP), which functionally mimics the subpopulation of neurons in a neural population. The DNP consists of two dynamic neural units which are coupled to function as excitatory and inhibitory neurons. The parallel architecture of the proposed neural model makes it very advantageous to apply to multi-variable control systems. The DNP is effectively implemented to coordinate different subsystems, both linear and nonlinear. The mathematical model and the learning and adaptive algorithm of the proposed neural model are described. A number of computer simulation studies are presented to demonstrate the effectiveness of the proposed model