Optimal control of the FitzHugh-Nagumo stochastic model with nonlinear diffusion (Q2234313)

The present work considers a stochastic optimal control problem governed by a variant of the (stochastic) FitzHugh-Nagumo model that contains a nonlinear diffusion term perturbed by multiplicative Brownian-type noise. The paper addresses the problem of existence and uniqueness of solutions of the state equations and the control problem. The paper commences with a motivation and a presentation of the model alongside the general notation and assumptions underpinning the mathematical analysis. The subsequent section presents the existence of the state equations. The paper continues with an analysis of the control problem. Basic existence results for the considered state equations are not applicable since the nonlinear polynomial perturbation associated with the ionic current is not \(m\)-accreative on the dual space associated with the diffusion operator. To address this issue, the original equation is transformed via a rescaling transformation to reduce the stochastic state equation to a random PDE. Based on this transformation, existence and uniqueness results for the state equation are provided. Likewise, a rescaling procedure is applied to the stochastic control problem to reduce it to a random optimal control problem. Ekeland's variational principle is considered to obtain a nearly optimal solution to the transformed control problem. Existence results of the controls and necessary optimality conditions are provided. The work concludes with proof of the bang-bang principle. The work is purely analytical. No numerical results are provided.