Solving Control Problems with Linear State Dynamics - A Practical User Guide

Abstract

In industrial applications, practitioners usually facea considerable complexity when optimizing operating strategiesunder uncertainty. Typical real-world problems arising in practiceare notoriously challenging from a computational viewpoint, requiring solutions to Markov Decision problems in high dimensions. In this work, we address a novel approach to obtain anapproximate solution to a certain class of problems, whose stateprocess follows a controlled linear dynamics. Our techniques isillustrated by an implementation within the statistical languageR, which we discuss by solving a typical problem arising in practice.