Electrical Engineering Department
National University of Colombia
A.A. No. 5885, Bogota, D.C.
Colombia, South America
One of the limitations of linearization theory that prevents its application in practical problems is the need for exact knowledge of the plant. In this paper, this requirement is eliminated, and input/output linearization is implemented training a Static Neural Network (SNN). This approach requires measurement of the full plant state and knowledge of the plant relative degree.
It is shown that a static network can synthesise the state feedback coefficients that linearize a nonlinear control affine plant, and that the stability of the linearizing close loop can be guaranteed if the autonomous plant is asymptotically stable and the neural network coefficients and the external input are bounded.
Keywords: input-output feedback linearization, static state feedback control, multilayer neural networks, closed loop linearization, stability of feedback linearization, chemotaxis
Received July 24, 1996; revised December 24, 1996.
Communicated by Jhing-Fa Wang.