Controlling the level of sparsity in MPC

DOI10.1016/J.SYSCONLE.2014.12.002zbMATH Open1307.93151arXiv1401.1369OpenAlexW2116901509MaRDI QIDQ2018573FDOQ2018573

Daniel Axehill

Publication date: 24 March 2015

Published in: Systems \& Control Letters (Search for Journal in Brave)

Abstract: In optimization routines used for on-line Model Predictive Control (MPC), linear systems of equations are usually solved in each iteration. This is true both for Active Set (AS) methods as well as for Interior Point (IP) methods, and for linear MPC as well as for nonlinear MPC and hybrid MPC. The main computational effort is spent while solving these linear systems of equations, and hence, it is of greatest interest to solve them efficiently. Classically, the optimization problem has been formulated in either of two different ways. One of them leading to a sparse linear system of equations involving relatively many variables to solve in each iteration and the other one leading to a dense linear system of equations involving relatively few variables. In this work, it is shown that it is possible not only to consider these two distinct choices of formulations. Instead it is shown that it is possible to create an entire family of formulations with different levels of sparsity and number of variables, and that this extra degree of freedom can be exploited to get even better performance with the software and hardware at hand. This result also provides a better answer to an often discussed question in MPC; should the sparse or dense formulation be used. In this work, it is shown that the answer to this question is that often none of these classical choices is the best choice, and that a better choice with a different level of sparsity actually can be found.

Full work available at URL: https://arxiv.org/abs/1401.1369

zbMATH Keywords

optimizationsparsitypredictive controlRiccati recursion

Mathematics Subject Classification ID

Linear systems in control theory (93C05)

Cites Work

• Title not available (Why is that?)
• Application of interior-point methods to model predictive control
• Array algorithms for H/sup ∞/ estimation
• A sparse and condensed QP formulation for predictive control of LTI systems
• A primal-dual interior-point method for robust optimal control of linear discrete-time systems
• Efficient solution of second order cone program for model predictive control

Cited In (4)

• \texttt{acados} -- a modular open-source framework for fast embedded optimal control
• Reducing the Overhead of MPC over a Large Population
• Recent advances in quadratic programming algorithms for nonlinear model predictive control
• Gauss-Newton Runge-Kutta integration for efficient discretization of optimal control problems with long horizons and least-squares costs

Uses Software

• fast_mpc

   Recommendations

• An alternative use of the Riccati recursion for efficient optimization 👍 👎
• A sparse and condensed QP formulation for predictive control of LTI systems 👍 👎
• Efficient MPC optimization using Pontryagin's minimum principle 👍 👎
• Newton projection with proportioning using iterative linear algebra for model predictive control with long prediction horizon 👍 👎
• An online active set strategy to overcome the limitations of explicit MPC 👍 👎

This page was built for publication: Controlling the level of sparsity in MPC

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2018573)