A multilayer recurrent neural network for real-time synthesis of linear-quadratic optimal control systems
Jun Wang, Guang Wu · 2002
A multilayer recurrent neural network is proposed for synthesizing linear-quadratic optimal control systems by means of solving algebraic matrix Riccati equations in real time. The proposed recurrent neural network consists of four bidirectionally connected layers. Each layer consists of an array of neurons. The proposed recurrent neural network is shown to be capable of solving algebraic Riccati equations and synthesizing linear-quadratic control systems in real time. The operating characteristics of the recurrent neural network and closed-loop control systems are also demonstrated through two illustrative examples.>