On Stability of Nonlinear Observers Based on Neural Networks

In this paper, the stability problem of neural network based observers/identifiers for nonlinear systems is revisited when nonlinear-in-parameter neural networks (NLPNN) are employed. The proposed approach is based on decomposing the neural network into two subsystems. The first subsystem (Subsystem 1) consists of the estimation error and output-layer weight error and the second subsystem (Subsystem 2) consists of the hidden-layer weight error. The key to this decomposition is that the hidden-layer weights appear in Subsystem 1, only as an argument of a sigmoidal function and its derivative which are both known to be bounded. This allows us to regard the Subsystem 1 as a linear-in-parameter neural network (LPNN) whose stability proof is more straightforward. Having shown the stability of the first subsystem, the stability of the second subsystem is also shown subsequently without the requirement of having the limiting assumptions of previous work. The bound on estimation error can be made arbitrarily small by proper selection of design parameters. The estimation scheme is then employed to estimate the state of flexible joint manipulators.

[1]  N. Hovakimyan,et al.  An adaptive observer design methodology for bounded nonlinear processes , 2002, Proceedings of the 41st IEEE Conference on Decision and Control, 2002..

[2]  Heidar Ali Talebi,et al.  A stable neural network-based identification scheme for nonlinear systems , 2003, Proceedings of the 2003 American Control Conference, 2003..

[3]  Heidar Ali Talebi,et al.  A NEURAL-NETWORK BASED OBSERVER FOR FLEXIBLE-JOINT MANIPULATORS , 2002 .

[4]  Heidar Ali Talebi,et al.  State estimation for flexible-joint manipulators using stable neural networks , 2003, Proceedings 2003 IEEE International Symposium on Computational Intelligence in Robotics and Automation. Computational Intelligence in Robotics and Automation for the New Millennium (Cat. No.03EX694).

[5]  A. Tornambè,et al.  An approximate observer for a class of nonlinear systems , 1989 .

[6]  B. Siciliano,et al.  An asymptotically stable joint PD controller for robot arms with flexible links under gravity , 1992, [1992] Proceedings of the 31st IEEE Conference on Decision and Control.

[7]  E. M. Hemerly,et al.  Neural adaptive observer for general nonlinear systems , 2000, Proceedings of the 2000 American Control Conference. ACC (IEEE Cat. No.00CH36334).

[8]  Weiping Li,et al.  Applied Nonlinear Control , 1991 .

[9]  A. Tornambè,et al.  High-gain observers in the state and estimation of robots having elastic joints , 1989 .

[10]  J.A.R. Vargas,et al.  Robust neural adaptive observer for MIMO nonlinear systems , 1999, IEEE SMC'99 Conference Proceedings. 1999 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.99CH37028).

[11]  R.V. Patel,et al.  A stable neural network observer with application to flexible-joint manipulators , 2002, Proceedings of the 9th International Conference on Neural Information Processing, 2002. ICONIP '02..

[12]  Robert Hecht-Nielsen,et al.  Theory of the backpropagation neural network , 1989, International 1989 Joint Conference on Neural Networks.

[13]  R. Marino,et al.  Adaptive observers with arbitrary exponential rate of convergence for nonlinear systems , 1995, IEEE Trans. Autom. Control..

[14]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[15]  A. Poznyak,et al.  Output trajectory tracking using dynamic neural networks , 2000, Proceedings of the 39th IEEE Conference on Decision and Control (Cat. No.00CH37187).

[16]  Mark W. Spong,et al.  Robot dynamics and control , 1989 .

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

[18]  F. L. Lewis,et al.  Nonlinear observer design using dynamic recurrent neural networks , 1996, Proceedings of 35th IEEE Conference on Decision and Control.