An adaptive backstepping based non-linear controller for artificial pancreas in type 1 diabetes patients

Abstract

Abstract Artificial pancreas enables closed loop automated control for blood glucose regulation in type 1 diabetic patients. Simple backstepping controller for regulation of blood glucose level has recently been proposed in the literature. In this research work, we have proposed backstepping based adaptive controller for integration in artificial pancreas. Controller design has been based on Bergman's minimal model which represents the dynamics of blood glucose-insulin system of human body. Glucose effectiveness factor has been treated as an unknown parameter and its value has been adapted using Lyapunov based adaptive backstepping control approach. Effects of meal disturbance, physical non-linearities and sensor noise have also been considered in the controller design. A complete mathematical derivation of the proposed nonlinear controller has been described and simulation results have been presented using Matlab/Simulink environment. Results indicate improvement in tracking response and overshoot/undershoot characteristics as compared to some recently developed techniques in literature. Proposed controller assures dynamic stability against disturbances and deviations in human body parameters. Practical implementation of the proposed controller can result in better artificial pancreas.