Alpaqa: A matrix-free solver for nonlinear MPC and large-scale nonconvex optimization

TL;DR

Alpaqa is an open-source C++ library implementing an augmented Lagrangian method with PANOC for large-scale nonconvex optimization, demonstrating improved performance in NMPC and benchmark tests.

Contribution

The paper introduces alpaqa, a new library that combines an augmented Lagrangian approach with enhanced PANOC algorithms for efficient large-scale nonconvex optimization.

Findings

Effective in NMPC applications

Improves convergence on CUTEst benchmarks

Open-source and easy to integrate

Abstract

This paper presents alpaqa, an open-source C++ implementation of an augmented Lagrangian method for nonconvex constrained numerical optimization, using the first-order PANOC algorithm as inner solver. The implementation is packaged as an easy-to-use library that can be used in C++ and Python. Furthermore, two improvements to the PANOC algorithm are proposed and their effectiveness is demonstrated in NMPC applications and on the CUTEst benchmarks for numerical optimization. The source code of the alpaqa library is available at https://github.com/kul-optec/alpaqa and binary packages can be installed from https://pypi.org/project/alpaqa .