LOW SENSITIVITY DESIGN OF NONLINEAR SYSTEMS USING LINEARIZED MODELS

Abstract

Low sensitivity linearized models ensure that the predictability and stability of nonlinearized models are achieved in a very assertive manner. The stability, which characterizes the trajectory in terms of equilibrium point, is of utmost importance. It should be noted that an equilibrium is considered asymptotically stable when there is a Lyapunov function present. One must not overlook the fact that the linear optimal state feedback for quadratic criteria, which do not include sensitivity, has the remarkable ability to automatically reduce sensitivity. It is an undeniable fact that the maximization of performance measures or the minimization of cost functions plays a crucial role in ensuring optimization. The primary focus of the research conducted was to significantly reduce the adverse impact of uncertainty on low sensitivity nonlinear models. To achieve this, a single perturbation technique was employed to convert the high-order system into a reduced model. This approach not only simplifies implementation but also effectively mitigates the impact of uncertainty. Additionally, the application of the KF estimator further enhances the reliability of the methodology. In this particular case, the static output feedback gain emerges as a more static and reliable output feedback control method that eliminates the need for an observer to estimate the states of system.