Neural Inverse Optimal Pinning Control of Output Trajectory Tracking for Uncertain Complex Networks with Nonidentical Nodes

Abstract

This chapter presents the development of a control scheme, we named as neural inverse optimal pinning control to achieve output trajectory tracking on uncertain complex networks with nonidentical nodes. A recurrent high order neural network is used to identify the unknown system dynamics of a small fraction of nodes (pinned ones) and by means of this neural model, an inverse optimal controller is designed to synchronize the whole network at an output desired reference. The proposed controller effectiveness is illustrated via simulations. The illustrative example is composed of a network of ten different chaotic nodes.