NMPC Policy Imitation via Regularized universal kriging
DOI:
https://doi.org/10.17979/ja-cea.2026.47.13770Keywords:
Predictive control, Data-based control, Learning for control, Nonlinear predictive control, Constrained controlAbstract
This work approximates offline the feedback law of an NMPC using universal kriging, avoiding online optimization. From state-action pairs, a standard UK policy and an ℓ1-regularized variant on the kriging weights are built to promote sparser and more interpretable interpolations. We demonstrate the effectiveness of the methodology using a simulated swing-up model of a torque-constrained inverted pendulum.
References
Bemporad, A., Muñoz de la Peña, D., 2009. Multiobjective model predictive control. Automatica 45 (12), 2823–2830.
Binder, M., Darivianakis, G., Eichler, A., Lygeros, J., 2019. Approximate explicit model predictive controller using Gaussian processes. In: 2019 IEEE 58th Conference on Decision and Control (CDC). pp. 841–846.
Carnerero, A., Ramirez, D., Lucia, S., Alamo, T., 2023. Prediction regions based on dissimilarity functions. ISA Transactions 139, 49–59.
Cressie, N., Wikle, C. K., 2011. Statistics for spatio-temporal data. John Wiley & Sons.
Cressie, N. A., 1993. Statistics for Spatial Data. John Wiley & Sons, Inc.
Fabiani, F., Goulart, P. J., 2023. Reliably-stabilizing piecewise-affine neural network controllers. IEEE Transactions on Automatic Control 68 (9), 5201–5215.
Ferramosca, A., Limon, D., Camacho, E. F., 2014. Economic MPC for a changing economic criterion for linear systems. IEEE Transactions on Automatic Control 59 (10), 2657–2667.
Hastie, T., Tibshirani, R.,Wainwright, M., 2015. Statistical learning with sparsity: the lasso and generalizations. Chapman & Hall/CRC.
He, D., Li, H., Du, H., 2021. Lexicographic multi-objective MPC for constrained nonlinear systems with changing objective prioritization. Automatica 125, 109433.
Karg, B., Alamo, T., Lucia, S., 2021. Probabilistic performance validation of deep learning-based robust NMPC controllers. International Journal of Robust and Nonlinear Control 31 (18), 8855–8876.
Krige, D., 1962. Effective pay limits for selective mining. Journal of the South African Institute of Mining and Metallurgy,, 345–362.
Limon, D., Ferramosca, A., Alvarado, I., Alamo, T., 2018. Nonlinear MPC for tracking piece-wise constant reference signals. IEEE Transactions on Automatic Control 63 (11), 3735–3750.
Maddalena, E., da S. Moraes, C., Waltrich, G., Jones, C., 2020. A neural network architecture to learn explicit MPC controllers from data. IFACPapersOnLine 53 (2), 11362–11367, 21st IFAC World Congress.
Mariethoz, G., Caers, J., 2014. Multiple-point geostatistics. John Wiley & Sons, Ltd.
Matheron, G., 1963. Principles of geostatistics. Economic geology 58, 1246–1266.
Matsui, H., Yamamoto, K., Yamakawa, Y., Jun. 2025. Sparse estimation in kriging for functional data. Stochastic Environmental Research and Risk Assessment 39, 2413–2425.
Moreno-Blazquez, C., Fele, F., Alamo, T., 2026. Accelerated kriging interpolation for real-time grid frequency forecasting. Sustainable Energy, Grids and Networks 47, 102403.
Muskinja, N., Tovornik, B., 2006. Swinging up and stabilization of a real inverted pendulum. IEEE transactions on industrial electronics 53 (2), 631–639.
Nadales, J. M., Carnerero, A. D., Moreno-Blazquez, C., Haes-Ellis, R. M., Limon, D., 2023. Learning-based NMPC on SoC platforms for real-time applications using parallel lipschitz interpolation. IFAC-PapersOnLine 56 (2), 6298–6303, 22nd IFAC World Congress.
Parisini, T., Zoppoli, R., 1995. A receding-horizon regulator for nonlinear systems and a neural approximation. Automatica 31 (10), 1443–1451.
Rawlings, J. B., Mayne, D. Q., Diehl, M. M., 2017. Model Predictive Control: Theory, Computation, and Design, 2nd Edition. Nob Hill Publishing, Madison, Wisconsin.
Richter, S., Jones, C. N., Morari, M., 2012. Computational complexity certification for real-time MPC with input constraints based on the fast gradient method. IEEE Transactions on Automatic Control 57 (6), 1391–1403.
Rose, A., Schaub, P., Findeisen, R., 2025. Safe and high-performance learning of model predictive control using kernel-based interpolation. In: 2025 American Control Conference (ACC). pp. 91–96.
Schabenberger, O., Gotway, C. A., 2005. Statistical methods for spatial data analysis. Chapman and Hall/CRC.
Stein, M. L., 1999. Interpolation of spatial data: some theory for kriging. Springer.
Tokmak, A., Fiedler, C., Zeilinger, M. N., Trimpe, S., K¨ohler, J., 2025. Automatic nonlinear MPC approximation with closed-loop guarantees. IEEE Transactions on Automatic Control 70 (10), 6388–6403.
Williams, C. K., Rasmussen, C. E., 2006. Gaussian processes for machine learning. Vol. 2. MIT press Cambridge, MA.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Carlos Moreno Blázquez, Filiberto Fele, Teodoro Álamo Cantarero
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
