Stochastic Control for Linear Systems Driven by Fractional Noises

From MaRDI portal





This article is concerned with the optimal control of stochastic linear systems driven by a fractional Brownian motion \(W_t^H\), where the Hurst parameter satisfies \(H\geq1/2\). The authors first give some results concerning the stochastic calculus for these processes; they consider both Itô-like and Stratonovich integrals (Itô-like integrals are constructed by means of the Wick calculus). Then they study three control problems where the state \(x\) satisfies an Itô or Stratonovich equation \[ dx_t=(A_tx_t+B_tu_t)dt +(C_t x_t+D_tu_t)dW_t^H \] involving the control \(u\). In the first and second problems, the state is one-dimensional, the equation is of Itô type, and they only consider controls of the form \(u_t=K_tx_t\). In the third problem, the state can be multidimensional and the equation is of Stratonovich type. These problems are difficult, and except for the first one, the existence of a solution is not proved; however, in each case, the authors relate solutions to an equation for the control; in particular, for the third problem, they obtain a Riccati equation which is a backward stochastic differential equation. Thus, though many questions remain open, this work is certainly an interesting step in the study of these control problems.




Cited in
(54)








This page was built for publication: Stochastic Control for Linear Systems Driven by Fractional Noises

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5317117)