Title:Schrödinger Approach to Optimal Control of Large-Size Populations

Abstract:Large-size populations consisting of a continuum of identical, non-cooperative and non-interacting agents with stochastic dynamics are useful in modeling various biological and engineered systems. This note addresses the problem of designing optimal state-feedback controllers for such systems which guarantee closed-loop stability of the stationary density of agents, in the case that individual agents have Langevin type passive dynamics. We represent the corresponding optimality system, which consists of coupled forward-backward PDEs as decoupled Schrödinger equations, by introducing a novel variable transform. Spectral properties of the Schrödinger operator which underlie the stability analysis are used to obtain explicit control design constraints. We show the deep connection between the nonlinear Schrödinger equation and mean field games for agents with nonlinear Langevin dynamics. Our interpretation of the Schrödinger potential as the cost function of a closely related optimal control problem motivates a quadrature based algorithm to compute the control.