Abstract
<title>Abstract</title> <p>This paper proposes an adaptive sliding mode control (ASMC) framework for Type 1 Diabetes blood glucose regulation, integrating Taylor-series-based online meal-disturbance estimation in a unified adaptive structure. The unknown disturbance amplitude A is estimated via a Lyapunov-based adaptive law using a Taylor-parameterized regressor without requiring prior knowledge of disturbance bounds. This estimate is embedded in the adaptive sliding mode control (ASMC) law, which achieves finite-time convergence, as proven by a strict Lyapunov analysis. A simulation validation across multiple patient profiles demonstrates that the ASMC reduces postprandial peaks by 60.6 % 270.58 - 106.74mg/dL, achieves 28.2, % faster settling time 102, min, vs, 142, min, and eliminates chattering with continuous insulin delivery. Theoretical convergence bounds are validated with errors of <5%. The proposed framework offers robust, computationally efficient, and clinically reliable glucose regulation for artificial pancreas systems.</p>