Title:Feedback Design for Devising Optimal Epidemic Control Policies

Abstract:For reliable epidemic monitoring and control, this paper proposes a feedback mechanism design to effectively cope with data and model uncertainties. Using past epidemiological data, we describe methods to estimate the parameters of general epidemic models. Because the data could be noisy, the estimated parameters may not be accurate. Therefore, under uncertain parameters and noisy measurements, we provide an observer design method for robust state estimation. Then, using the estimated model and state, we devise optimal control policies by minimizing a predicted cost functional. Finally, the effectiveness of the proposed method is demonstrated through its implementation on a modified SIR epidemic model.