SDRE 방식을 이용한 HIV 모델의 최적 제어 이론

초록

We present and use techniques and ideas from control theory to design and synthesize nonlinear feedback control-based treatment regimes for HIV. We demonstrate through numerical simulations that by using a ``target tracking'' approach, suboptimal feedback-based treatment strategies can be designed to move the state of the system from an ``unhealthy'' state (high virus load and low immune response) to a ``healthy'' one (with low viral load and high immune effector levels). An important advantage of this drug regimen design is that once the viral load is controlled to very low levels, the drug dosage can be reduced or completely terminated. Consequently, long term pharmaceutical side effects could also be reduced. Thus, this approach suggests that by anticipating and responding to the disease progression, dynamic feedback strategies such as those designed in this work could lead to long-term control of HIV after discontinuation of therapy.