On a new approach to the design of tracking controllers for nonlinear dynamical systems
暂无分享,去创建一个
The nonlinear autoregressive moving-average (NARMA) model is now well established for identifying nonlinear systems using input-output data. In this paper a method of designing a controller based on the NARMA model is proposed. The controller is shown to be the solution to an implicit algebraic equation, so that the controller can be determined without explicit use of past values of inputs and outputs as in traditional methods. Simulation studies are presented to demonstrate that the method proposed is significantly better than other methods currently in use. The resulting controller provides a satisfactory starting point for online adaptation in practical applications.
[1] Kumpati S. Narendra,et al. Identification and control of dynamical systems using neural networks , 1990, IEEE Trans. Neural Networks.
[2] Robert M. Sanner,et al. Gaussian Networks for Direct Adaptive Control , 1991, 1991 American Control Conference.
[3] Frank L. Lewis,et al. Multilayer neural-net robot controller with guaranteed tracking performance , 1996, IEEE Trans. Neural Networks.
[4] Marios M. Polycarpou,et al. Stable adaptive neural control scheme for nonlinear systems , 1996, IEEE Trans. Autom. Control..