Full-Order Safety for Nonlinear Reduced-Order MPC
Abstract
Reduced-order models (ROMs) make model predictive control (MPC) of high-dimensional nonlinear systems tractable, but the resulting modeling error can lead to constraint violations on the real system. We introduce CoRE-MPC, a robust MPC formulation that plans with a nonlinear ROM and guarantees recursive feasibility and constraint satisfaction for the full-order system. It can be applied with a broad class of nonlinear model reduction techniques and the resulting MPC does not require evaluating the full-order model online, thus ensuring computational efficiency. The key insight is in how the ROM affects the dynamics of the prediction error. We account for this with two error radii that are propagated alongside the ROM and provide a guaranteed bound on the prediction error. The approach is validated in simulation on an 80-state oscillator chain, a Burgers equation, and a 9,768-state finite-element soft robot, where a two-state ROM with data-driven bounds keeps all 1,628 nodes clear of a wall at a median solve time below 1 ms for a 20 ms control period. Across these examples, CoRE-MPC is 13 to 61 times faster than full-order robust MPC at only 27 to 35% higher closed-loop cost, whereas existing certified schemes are either infeasible or at least twice as conservative. Code: [released upon publication].
Related
- Reduced-Order Model Characterization of Nonlinear Sustained Oscillations
- Lifting Safety Through Reduced-Order Model Architectures: An Invariant Manifold Approach
- A Nonlinear Physics-based Reduced Order Model with Convolutional-based Operator Compression
- Stochastic realisers of non-degenerate full-degree Type II reduced Ito polynomials
- Modified wave operators for nonlinear Schrödinger equations in the full subcritical long range regime
BibTeX
@misc{pabon2026fullorder,
title = {{Full-Order Safety for Nonlinear Reduced-Order MPC}},
author = {Luis Pabon and Johannes Köhler and Revant Kasichainula and Marco Pavone},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.19250},
eprint = {2610.19250},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.19250},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Full-Order Safety for Nonlinear Reduced-Order MPC
AU - Luis Pabon
AU - Johannes Köhler
AU - Revant Kasichainula
AU - Marco Pavone
PY - 2026
DA - 2026/10/13
PB - arXiv
DO - 10.48550/arxiv.2610.19250
UR - https://arxiv.org/abs/2610.19250
AB - Reduced-order models (ROMs) make model predictive control (MPC) of high-dimensional nonlinear systems tractable, but the resulting modeling error can lead to constraint violations on the real system. We introduce CoRE-MPC, a robust MPC formulation that plans with a nonlinear ROM and guarantees recursive feasibility and constraint satisfaction for the full-order system. It can be applied with a broad class of nonlinear model reduction techniques and the resulting MPC does not require evaluating the full-order model online, thus ensuring computational efficiency. The key insight is in how the ROM affects the dynamics of the prediction error. We account for this with two error radii that are propagated alongside the ROM and provide a guaranteed bound on the prediction error. The approach is validated in simulation on an 80-state oscillator chain, a Burgers equation, and a 9,768-state finite-element soft robot, where a two-state ROM with data-driven bounds keeps all 1,628 nodes clear of a wall at a median solve time below 1 ms for a 20 ms control period. Across these examples, CoRE-MPC is 13 to 61 times faster than full-order robust MPC at only 27 to 35% higher closed-loop cost, whereas existing certified schemes are either infeasible or at least twice as conservative. Code: [released upon publication].
ER -