Performance Bounds of Model Predictive Control for Unconstrained and Constrained Linear Quadratic Problems and Beyond

TL;DR

This paper analyzes the performance bounds of model predictive control (MPC) for linear quadratic problems, providing insights into how to select terminal costs and constraints to improve feasibility and reduce performance loss.

Contribution

The authors derive new performance bounds for MPC applied to LQ problems and propose alternative terminal cost and constraint choices independent of Riccati solutions.

Findings

MPC can achieve near-optimal performance with proper terminal choices.

New terminal strategies expand feasible regions without significant cost increase.

Performance bounds guide better MPC design for constrained and unconstrained problems.

Abstract

We study unconstrained and constrained linear quadratic problems and investigate the suboptimality of the model predictive control (MPC) method applied to such problems. Considering MPC as an approximate scheme for solving the related fixed point equations, we derive performance bounds for the closed-loop system under MPC. Our analysis, as well as numerical examples, suggests new ways of choosing the terminal cost and terminal constraints, which are \emph{not} related to the solution of the Riccati equation of the original problem. The resulting method can have a larger feasible region, and cause hardly any loss of performance in terms of the closed-loop cost over an infinite horizon.