Optimal control of nonlinear systems using RBF neural network and adaptive extended Kalman filter

This paper presents a nonlinear optimal control technique based on approximating the solution to the Hamilton-Jacobi-Bellman (HJB) equation. The HJB solution (value function) is approximated as the output of a radial basis function neural network (RBFNN) with unknown parameters (weights, centers, and widths) whose inputs are the system's states. The problem of solving the HJB equation is therefore converted to estimating the parameters of the RBFNN. The RBFNN's parameters estimation is then recognized as an associated state estimation problem. An adaptive extended Kalman filter (AEKF) algorithm is developed for estimating the associated states (parameters) of the RBFNN. Numerical examples illustrate the merits of the proposed approach.

[1]  Panagiotis Tsiotras,et al.  Sub-optimal feedback control using a successive wavelet-Galerkin algorithm , 2003, Proceedings of the 2003 American Control Conference, 2003..

[2]  Radhakant Padhi,et al.  Continuous-Time Single Network Adaptive Critic for Regulator Design of Nonlinear Control Affine Systems , 2008 .

[3]  M. E. Ahmed,et al.  Neural controllers for nonlinear state feedback L/sub 2/-gain control , 2000 .

[4]  George G. Lendaris,et al.  Adaptive dynamic programming , 2002, IEEE Trans. Syst. Man Cybern. Part C.

[5]  Chen-Chung Liu,et al.  Adaptively controlling nonlinear continuous-time systems using multilayer neural networks , 1994, IEEE Trans. Autom. Control..

[6]  George N. Saridis,et al.  An Approximation Theory of Optimal Control for Trainable Manipulators , 1979, IEEE Transactions on Systems, Man, and Cybernetics.

[7]  William L. Garrard,et al.  Design of nonlinear automatic flight control systems , 1977, Autom..

[8]  Dan Simon,et al.  Training radial basis neural networks with the extended Kalman filter , 2002, Neurocomputing.

[9]  Chaouki T. Abdallah,et al.  Linear Quadratic Control: An Introduction , 2000 .

[10]  W. Garrard,et al.  Nonlinear feedback control of highly manoeuvrable aircraft , 1992 .

[11]  Behnaam Aazhang,et al.  Dynamic Neural Networks for Modeling and Control of Nonlinear Systems , 2003, Intell. Autom. Soft Comput..

[12]  Konrad Reif,et al.  An EKF-Based Nonlinear Observer with a Prescribed Degree of Stability , 1998, Autom..

[13]  F. Lewis,et al.  A Hamilton-Jacobi setup for constrained neural network control , 2003, Proceedings of the 2003 IEEE International Symposium on Intelligent Control.

[14]  J. Cloutier,et al.  Control designs for the nonlinear benchmark problem via the state-dependent Riccati equation method , 1998 .

[15]  Randal W. Beard,et al.  Galerkin approximations of the generalized Hamilton-Jacobi-Bellman equation , 1997, Autom..