Novel algebraic meal disturbance estimation based adaptive robust control design for blood glucose regulation in type 1 diabetes patients

Observer‐based control for plasma glucose regulation in type 1 diabetes mellitus patients with unknown input delay

“…33 This is a common approach adopted in the literature in the context of automatic control. 7,11,32,[34][35][36] In this sense, the control signal u is the additional insulin bolus to control glucose increases associated with food intake or to correct elevated levels.…”

“…Using the comparison lemma of, 38 the observation error x (18) converges exponentially to a region close to the origin, then the state x also converges. Now, Λ c is a bilinear matrix inequality; therefore, by introducing an appropriate non-singular transformation to (35), that is,…”

“…However, since the objective here is to address the glucose increments caused by unplanned meals, it is assumed that the patient is a stable diabetic individual controlled by subcutaneous infusion, and the effect of the basal insulin administration () is considered in the model by setting to an appropriate value 33 . This is a common approach adopted in the literature in the context of automatic control 7,11,32,34–36 . In this sense, the control signal is the additional insulin bolus to control glucose increases associated with food intake or to correct elevated levels.…”

Design of dual hormone blood glucose therapy and comparison with single hormone using MPC algorithm

“…Few research studies examined by using disturbances such as meal and exercise, and a comparative study was performed on open-and closed-loop system clinically by Russell et al [27], Reiter et al [29] and Haidar et al [30]. Recent research on blood glucose regulation and modelling dynamics of plant model represented using robust control algorithm with Bergman model using sliding mode control integrating algebraic meal as disturbance [31], other research developed Fuzzy-based algorithm with identifier and are tuned using backpropagation [32], and in silico study for development was proposed by developing an adaptive controller with intrapatient variability handling structure [33], modelling of plant model and then simulation study was adapted by Imamovi et al [34], with consideration of the realistic situation of time-varying model a robust observer-based adaptive controller is proposed on a non-linear uncertain model. For the first time, they have integrated the robustness and adaptive philosophy claiming better regulation [35].…”

New Contributions Towards Meal and Exercise Announcement-Free Artificial Pancreas Systems

“…State and disturbance observers are a suitable real-time alternative. Although its accuracy is lower than the gold standard method, they have been successfully applied for either meal detection (Kölle et al 2017;Ramkissoon et al 2018;Chen et al 2019) or compensation of unannounced meals (Kovacs et al 2019;Ullah et al 2020;Sanz et al 2020). The Kalman Filter (KF) is the prevalent observer option in the artificial pancreas literature; however, tuning the covariance matrix is not straightforward.…”