Wireless neural control network for nonlinear systems

This paper is concerned with the distributed control problem over a wireless neural control network (WNCN) for multi-input and multi-output (MIMO) nonlinear systems. Our approach treats the WNCN as a fully distributed controller. Specifically, a cooperative control strategy for every wireless neuron nodes in the WNCN to follow, where there is only wireless communications interlinks among neighbor neuron nodes. This imposed the sparsity constraints to the network topology. It means the WNCN has considerably fewer weights than fully connected neural networks, which is conducive to the solution of industrial real-time control problem. The stability analysis of the closed-loop control system is derived from the discrete Lyapunov stability theorem according to the learning rate parameter. Finally, a simulation example illustrates the effectiveness of the proposed method.