A new robust nonlinear control algorithm for the regulation of blood glucose in diabetic patients

Diabetes mellitus is a growing health problem worldwide. Especially, the patients with Type 1 diabetes need strict glycemic control because they have deficiency of insulin production. Today, the aim of the researchers is to develop a fully automated closed loop control system (i. e., artificial pancreas which is capable of continuous glucose sensing and regulating) that replicates the function of pancreatic beta cells. This equipment is supposed to be capable of maintaining normal blood glucose level on a desired level (4-6.2 mmol/L) and taking all factors affecting glucose concentration into account without inducing hypoglycemia and hyperglycemia. In this paper, control of the Bergman minimal mathematical model to regulate blood glucose level is discussed. The control problem is complicated by the fact that the model is considered to be uncertain. Furthermore, only the glucose level is assumed to be available for control development and other states such as insulin concentration and plasma insulin concentration are assumed to be unavailable for measurement. A novel robust controller is proposed that guarantees practical tracking of a desired glucose concentration. © 2012 IEEE.TUBITAK; Ministry of Science, Industry and Technology of Turkey (grant 1402.TGSD.2012