Blood Glucose Regulation with Hypoglycemic Avoidance: An Interval-Predictor Approach

The manuscript is devoted to blood glucose regulation in type 1 diabetic patients using a robust switching control. The controller relies on an interval–based predictor that estimates the state considering estate and input constraints. The proposed controller can deal with the uncertainties in the patient parameters and external perturbations such as unannounced food intake. The closed–loop stability analysis relies on the Lyapunov function approach. The blood glucose is stabilized around the desired value while the state and control trajectories remain within the desired constraints. The feasibility of the approach was validated in simulations using a cohort of four adult patients.