Observer-Based Output Feedback Linearization Control with Application to HIV Dynamics

Abstract

This paper presents the feedback linearization control of HIV infection. A multi-input multi-output (MIMO) dynamic nonlinear HIV infection model for this purpose has been used. For this purpose, three widely used drugs are considered. A Luenberger-like nonlinear observer (LNO) is designed for estimation of unavailable states. To minimize the side effects of drugs, the concentration of ZDV which has the highest side effect is fixed to a minimum value and the external controllers parameters are obtained by maximizing an objective function. In the control design, limitations on drug consumption and unavailability of all states are taken into account. The closed-loop stability has been established in the presence of the observer dynamics and input limitations. Robustness of the proposed treatment strategy against model parameter uncertainties, noise, or sparse measurements has been demonstrated through simulation study. Finally, by including pharmacological models, an impulsive control strategy for HIV treatment has been considered.

Observer-Based Impulsive Controller Design for Treatment of Hepatitis C Disease

“…The unavailability of system states because of the technical limitations or high prices of measuring devices is a problem in the controller design. Observers have been designed in order to estimate the unmeasured states of linear and nonlinear systems, − and they have also been used for several biological applications. − Impulsive observers have been proposed to estimate the unknown states of the system when the output measurements are only available at special discrete times. , Since, in the HCV treatment process, the viral load is measured at certain sampling times, an impulsive observer should be designed to estimate the unknown states for implementation of the treatment strategy. For this purpose, the impulsive Luengerger-like observer, which has been extensively utilized for the nonlinear systems, has been chosen.…”

Adaptive Control Strategy for Treatment of Hepatitis C Infection

“…Control theories have been extensively utilized in biological systems, , for example for HIV − and HCV treatments. − Chakrabarty and Joshi presented a treatment strategy for combination therapy based on offline optimal control strategy. An objective function has been established in order to minimize the viral load and side effects of drugs.…”

A Moving Horizon Estimation Algorithm Based on Carleman Approximation: Design and Stability Analysis

“…However, knowledge of these state variables is critical in order to effectively control the system and reach the desirable output performance. In the literature, there are many observer design techniques to estimate the unknown system states using measurements of the input and output trajectories. , These techniques range from Luenberger observer for linear systems to more complex designs for nonlinear systems such as extended Luenberger observer, − linearization by output injection, high-gain observer, and reduced-order observer. − The state estimation becomes even harder when the measurements are corrupted with noise. The moving horizon estimation (MHE) method is a state estimation method which accounts for noise and/or disturbance on system output and system transition .…”