Approximate Solution Methods for the Average Reward Criterion in Optimal Tracking Control of Linear Systems

TL;DR

This paper develops approximate control methods for linear systems optimizing the average reward, deriving the value function and policy, and applying Model Predictive Control for practical use.

Contribution

It introduces an approximate solution framework for average reward optimal control in linear systems using Model Predictive Control.

Findings

Derived the value function and optimal policy for the average reward criterion.

Proposed an MPC-based approximate solution for practical implementation.

Demonstrated the effectiveness of the approach in linear system control.

Abstract

This paper studies optimal control under the average-reward/cost criterion for deterministic linear systems. We derive the value function and optimal policy, and propose an approximate solution using Model Predictive Control to enable practical implementation.