Gain-scheduled control of a system with input constraint by suppression of input derivatives

Abstract

The purpose of this study is to design a gain-scheduled feedback compensator for a system constrained on the input. Description of the input saturation by the hyperbolic tangential function transfers the system description to a form of a linear parameter-dependent system. This formulation allows applying the gain-scheduling control synthesis via linear matrix inequalities (LMI). In order to avoid the windup phenomena caused by input limitation suppression of the input derivatives is employed. While the input derivatives could not be taken into account in the previous studies because of using the discontinuous functions as the input limitations, in this study it is available to take account of the input derivatives in the controller design. This means that the designed controller includes integrators and constructs a type-one control system. Consideration of input derivatives may mitigate sudden change of the input even if there is no feedback loop of error between input and output of the nonlinear function. Simulation results show usefulness of the proposed design method.