An input-output based robust stabilization criterion for neural-network control of nonlinear systems

A stabilization method based on the input-output conicity criterion is presented. Conventional learning algorithms are applied to adjust the controller dynamics, and robust stability of the closed-loop system is guaranteed by modifying the training patterns which yield unstable behavior. The methodology developed expands the class of nonlinear systems to be controlled using neural control schemes, so that the stabilization of a broad class of neural-network-based control systems, even with unknown dynamics, is assured. Straightforwardness in the application of this method is evident in contrast to the Lyapunov function approach.

[1]  T. Georgiou,et al.  Robustness analysis of nonlinear feedback systems: an input-output approach , 1997, IEEE Trans. Autom. Control..

[2]  C. Desoer,et al.  Feedback Systems: Input-Output Properties , 1975 .

[3]  Johan A. K. Suykens,et al.  NLq Theory: A Neural Control Framework with Global Asymptotic Stability Criteria , 1997, Neural Networks.

[4]  O. Hijab.  Input/Output Properties , 1987 .

[5]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1989, Math. Control. Signals Syst..

[6]  Eli Tzirkel-Hancock,et al.  Stable control of nonlinear systems using neural networks , 1992 .

[7]  A. Barreiro,et al.  Robust stability of fuzzy control systems based on conicity conditions , 1999, Autom..

[8]  Yutaka Fukuoka,et al.  A modified back-propagation method to avoid false local minima , 1998, Neural Networks.

[9]  R.R. Mohler,et al.  Stability and robustness of multivariable feedback systems , 1981, Proceedings of the IEEE.

[10]  M. Vidyasagar,et al.  New relationships between input-output and Lyapunov stability , 1982 .

[11]  Johan A. K. Suykens,et al.  Lur'e systems with multilayer perceptron and recurrent neural networks: absolute stability and dissipativity , 1999, IEEE Trans. Autom. Control..

[12]  D. Bernstein,et al.  Explicit construction of quadratic lyapunov functions for the small gain, positivity, circle, and popov theorems and their application to robust stability. part II: Discrete-time theory , 1993 .

[13]  M. Vidyasagar,et al.  Nonlinear systems analysis (2nd ed.) , 1993 .

[14]  Andrew R. Teel.  On graphs, conic relations and input-output stability of nonlinear feedback systems , 1995, Proceedings of 1995 34th IEEE Conference on Decision and Control.

[15]  C. A. Desoer,et al.  Nonlinear Systems Analysis , 1978 .

[16]  Kurt Hornik,et al.  Degree of Approximation Results for Feedforward Networks Approximating Unknown Mappings and Their Derivatives , 1994, Neural Computation.

[17]  Roberto Battiti,et al.  First- and Second-Order Methods for Learning: Between Steepest Descent and Newton's Method , 1992, Neural Computation.

[18]  S. Arik,et al.  Stability analysis of dynamical neural networks , 1997 .

[19]  A. García-Cerezo,et al.  Robust design of rule-based fuzzy controllers , 1995 .