Data-driven policy iteration algorithm for continuous-time stochastic linear-quadratic optimal control problems

TL;DR

This paper introduces a data-driven policy iteration algorithm for solving infinite-horizon stochastic linear-quadratic control problems without prior knowledge of system matrices, demonstrating effectiveness through simulation.

Contribution

It proposes a novel data-driven approach to solve stochastic LQ problems by iteratively approximating the Riccati equation without system model knowledge.

Findings

Algorithm successfully approximates the solution to the SARE.

Simulation confirms the algorithm's effectiveness and applicability.

Method enables model-free control in stochastic LQ problems.

Abstract

This paper studies a continuous-time stochastic linear-quadratic (SLQ) optimal control problem on infinite-horizon. A data-driven policy iteration algorithm is proposed to solve the SLQ problem. Without knowing three system coefficient matrices, this algorithm uses the collected data to iteratively approximate a solution of the corresponding stochastic algebraic Riccati equation (SARE). A simulation example is provided to illustrate the effectiveness and applicability of the algorithm.