using Rl an agent learns how much insulin to pump. to model the patient we used a differential equation. which is diabete1.m more information about the parameters and the patient model could be found in the original paper which this project is based on. "Using Sliding Mode Controller and Eligibility Traces for Controlling the Blood Glucose in Diabetic Patients at the Presence of Fault" by A. Noori, M. A. Sadrniaa, M. B. Naghibi Sistani they used "Eligibility Traces" , but instead we used Q-learning.