Integral Backstepping Based Automated Control of Blood Glucose in Diabetes Mellitus Type 1 Patients

Abstract

Diabetes Mellitus Type 1 happens when our immune system destroys beta cells in our pancreas due to which it fails to produce enough insulin; a hormone which allows sugar/glucose to enter in its cells in order to produce energy. To cope with failure of pancreas, artificial ones are used to inject the required amount of insulin in the body. Controllers are used for automatic balancing of blood glucose-insulin level. Bergman's minimal model (BMM) is a physiologically verified model representing this phenomenon. In a recent research BMM is extended to more generic form with an extended state of the system, dealing with the disturbance to the blood glucose level caused by meal intake during medication. In this research paper, we have used BMM along with its extended model and proposed three nonlinear controllers: Integral Backstepping (IBS) Controller, Backstepping (BS) Controller and Fuzzy Logic Controller (FLC), for the automatic stabilization of the blood glucose level in Diabetes Mellitus Type 1 patients. The integral action is integrated with Backstepping technique; resulting in reducing steady-state error by significant amount. A mathematical analysis has been done to prove the asymptotic stability of the proposed controllers for the both models using Lyapunov theory. For showing the tracking behavior of the proposed controllers with their respective models to the desired blood-glucose output, simulation results have been performed and discussed using MATLAB/Simulink. Comparison results of the both systems show that the proposed controllers performs far better than the ones given in the literature.

Keywords

• Bergman’s minimal model (BMM)
• integral backstepping (IBS)
• backstepping (BS)
• fuzzy logic controller (FLC)