An algorithm for the structural analysis of state space: synthesis of nonlinear observers

The problem addressed is the linearization of multi-input multi-output (MIMO) nonlinear systems by a generalized state coordinates transformation and generalized input–output injection, in order to design an observer. This observer will have linear error dynamics. The goal is to bring together two observers design approaches: a structural one and a numerical one. Necessary and sufficient conditions for the existence of a linearizing generalized state transformation are obtained by an algebraic way and without computing the input–output differential equations. The main result tests integrability conditions of differential one-forms derived from the state space representation and is applicable to a large subclass of nonlinear systems. Copyright © 2001 John Wiley & Sons, Ltd.

[1]  Jessy W. Grizzle,et al.  Interpolation and numerical differentiation for observer design , 1994, Proceedings of 1994 American Control Conference - ACC '94.

[2]  T. Ph. Proychev,et al.  Transformation of nonlinear systems in observer canonical form with reduced dependency on derivatives of the input , 1993, Autom..

[3]  R. Marino,et al.  Dynamic output feedback linearization and global stabilization , 1991 .

[4]  J. Gauthier,et al.  Bilinearization up to output injection , 1988 .

[5]  A. Krener,et al.  Nonlinear observers with linearizable error dynamics , 1985 .

[6]  Darrell Williamson,et al.  Observation of bilinear systems with application to biological control , 1977, Autom..

[7]  S. Diop,et al.  On continuous time differentiation observers , 1999, 1999 European Control Conference (ECC).

[8]  F. Plestan,et al.  Algebraic Conditions for Linearization by Input-Output Injection , 1995 .

[9]  F. Plestan,et al.  Linearization by Generalized Input-Output Injection and Synthesis of Observers , 1996 .

[10]  Michel Fliess,et al.  Generalized controller canonical form for linear and nonlinear dynamics , 1990 .

[11]  Franck Plestan,et al.  Linearization by generalized input-output injection , 1997 .

[12]  C. Moog,et al.  New algebraic-geometric conditions for the linearization by input-output injection , 1996, IEEE Trans. Autom. Control..

[13]  M. Zeitz,et al.  Extended Luenberger observer for non-linear multivariable systems , 1988 .

[14]  H. Keller.  Non-linear observer design by transformation into a generalized observer canonical form , 1987 .

[15]  Xiaohua Xia,et al.  Input output linearization of nonlinear systems by output feedback , 1996, Proceedings of 35th IEEE Conference on Decision and Control.

[16]  Franck Plestan,et al.  Synthesis of nonlinear observers via structural analysis and numerical differentiation , 1999, 1999 European Control Conference (ECC).

[17]  X. Xia,et al.  Nonlinear observer design by observer error linearization , 1989 .

[18]  A. V. D. Wegen.  Representing a nonlinear state space system as a set of higher-order differential equations in the inputs and outputs , 1989 .

[19]  A. Glumineau,et al.  Further results on linearization of nonlinear systems by input-output injection , 1997, Proceedings of the 36th IEEE Conference on Decision and Control.

[20]  John Chiasson,et al.  Nonlinear Differential-Geometric Techniques for Control of a Series DC Motor , 1993, 1993 American Control Conference.

[21]  F. Plestan,et al.  An Observer for a One Flexible Joint Robot by an Algebraic Method , 1994 .