Error control in model following control systems using constant error model following

Abstract

This thesis describes the development of a new method for controlling the error in model following control systems. The treatment is for first order, linear or nonlinear, time-varying or time invariant systems with additive (linear) control. The errors controlled are assumed to have arisen from external disturbances or from differences in the initial conditions of the plant and the model. The theory introduced is called constant error model following control. This paper describes the theory as an outgrowth of attempts to control a plant by feedback of an error between the plant and model with the error specified to be constant. From this, it is shown that a model may be followed with arbitrary error. The central result is that, given some error, one can find another model (control model) which if followed with this arbitrary error, will guide the plant state trajectory back to that of the model.