Fully tuned radial basis function neural networks for flight control (Q2761432)

Recently, neural networks (NN) based control strategies have attracted much attention. Among a variety of network structures, radial basis function (RBF) NN have been studied intensively. The reviewed book is the first monograph devoted to fully tuned RBF NN for a very special application, namely for a flight control problem. Chapter 1 presents a detailed review of the RBF NN in the field of nonlinear adaptive control, with an emphasis on aircraft flight control applications. The rest of the book has been divided into two parts. Part I is devoted to indirect adaptive control. In Chapter 2 different sequential learning algorithms are evaluated for nonlinear system identification. Here the Lyapunov method is used. Chapter 3 presents the minimal resource allocation network (MRAN) for the identification of nonlinear systems. Besides, a new algorithm, called the extended MRAN (EMRAN), is developed to increase the on-line learning speed. The last chapter of Part I describes the performance of the indirect adaptive control.