Optimal Robust Controller Design for Uncertain Linear Glucose System

Abstract

In this paper, insulin injection process in which the level of glucose is monitored and corrected using optimal robust controller. The Bergman minimal model with unknown parameters was adopted to simulate a diabetes mellitus patient. An exogenous disturbance is considered as a meal effect on the patient. Optimal control approach to robust control design is suggested for glucose problem. The goal is to design a state feedback to stabilize the glucose system under uncertainty and disturbance. Linearized model of glucose system is considered with nonlinearity is lumped as a matched and mismatched uncertainties. Comparison in performance is made between the robust optimal controllers with classical optimal controller. The simulated results showed that the robust optimal controller outperforms the classical optimal controller for the considered cases. Also, the disturbance rejection capability of both controllers has been examined, and the results showed that the robust optimal controller has better rejection characteristics than its counterpart.