Efficient Representation and Approximation of Model Predictive Control Laws via Deep Learning
Open accessThe DOI matches its Crossref record; 46 of the 46 references it cites were found in a registry or an index.
Abstract
We show that artificial neural networks with rectifier units as activation functions can exactly represent the piecewise affine function that results from the formulation of model predictive control (MPC) of linear time-invariant systems. The choice of deep neural networks is particularly interesting as they can represent exponentially many more affine regions compared to networks with only one hidden layer. We provide theoretical bounds on the minimum number of hidden layers and neurons per layer that a neural network should have to exactly represent a given MPC law. The proposed approach has a strong potential as an approximation method of predictive control laws, leading to a better approximation quality and significantly smaller memory requirements than previous approaches, as we illustrate via simulation examples. We also suggest different alternatives to correct or quantify the approximation error. Since the online evaluation of neural networks is extremely simple, the approximated controllers can be deployed on low-power embedded devices with small storage capacity, enabling the implementation of advanced decision-making strategies for complex cyber-physical systems with limited computing capabilities.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Cortes, C., & Vapnik, V. (1995) · Machine Learning · doi:10.1007/bf00994018 · Cited by 59,421
Boyd, S., & Vandenberghe, L. (2004) · doi:10.1017/cbo9780511804441 · Cited by 56,601
Boyd, S., Parikh, N., Chu, E., Peleato, B., & Eckstein, J. (2011) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000016 · Cited by 17,075
Mastering the game of Go with deep neural networks and tree search
Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., van den Driessche, G., Schrittwieser, J., Antonoglou, I., Panneershelvam, V., Lanctot, M., et al. (2016) · Nature · doi:10.1038/nature16961 · Cited by 15,612
Boyd, S. (2010) · now publishers, Inc. eBooks · doi:10.1561/9781601984616 · Cited by 12,677
Constrained model predictive control: Stability and optimality
Mayne, D. Q., Rawlings, J. B., Rao, C. V., & Scokaert, P. O. M. (2000) · Automatica · doi:10.1016/s0005-1098(99)00214-9 · Cited by 8,736
Probability Inequalities for sums of Bounded Random Variables
Hoeffding, W. (1994) · Springer Series in Statistics · doi:10.1007/978-1-4612-0865-5_26 · Cited by 6,666
Model predictive control: Theory and practice—A survey
García, C. E., Prett, D. M., & Morari, M. (1989) · Automatica · doi:10.1016/0005-1098(89)90002-2 · Cited by 5,284
A survey of industrial model predictive control technology
Qin, S. J., & Badgwell, T. A. (2003) · Control Engineering Practice · doi:10.1016/s0967-0661(02)00186-7 · Cited by 5,099
(2008) · Cited by 4,811
The explicit linear quadratic regulator for constrained systems
Bemporad, A., Morari, M., Dua, V., & Pistikopoulos, E. N. (2002) · Automatica · doi:10.1016/s0005-1098(01)00174-1 · Cited by 3,260
Universal approximation bounds for superpositions of a sigmoidal function
Barron, A. R. (1993) · IEEE Transactions on Information Theory · doi:10.1109/18.256500 · Cited by 2,804
Predictive Control for Linear and Hybrid Systems
Borrelli, F., Bemporad, A., & Morari, M. (2017) · Cambridge University Press eBooks · doi:10.1017/9781139061759 · Cited by 1,443
Fast Model Predictive Control Using Online Optimization
Wang, Y., & Boyd, S. (2010) · IEEE Transactions on Control Systems Technology · doi:10.1109/tcst.2009.2017934 · Cited by 1,372
Herceg, M., Kvasnica, M., Jones, C. N., & Morari, M. (2013) · 2013 European Control Conference (ECC) · doi:10.23919/ecc.2013.6669862 · Cited by 1,178
Srivastava, R. K., Greff, K., & Schmidhuber, J. (2015) · arXiv · doi:10.48550/arxiv.1505.00387 · Cited by 912
On the Number of Linear Regions of Deep Neural Networks
Montúfar, G., Pascanu, R., Cho, K., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1402.1869 · Cited by 840
CVXGEN: a code generator for embedded convex optimization
Mattingley, J., & Boyd, S. (2012) · Optimization and Engineering · doi:10.1007/s11081-011-9176-9 · Cited by 790
Model predictive control with signal temporal logic specifications
Raman, V., Donzé, A., Maasoumy, M., Murray, R. M., Sangiovanni-Vincentelli, A. L., & Seshia, S. A. (2014) · IEEE Conference on Decision and Control (CDC) · doi:10.1109/cdc.2014.7039363 · Cited by 540
A Survey on Explicit Model Predictive Control
Alessio, A., & Bemporad, A. (2009) · Lecture Notes in Control and Information Sciences · doi:10.1007/978-3-642-01094-1_29 · Cited by 511
Liu, L., Liu, Y.-J., & Tong, S. (2019) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2018.2828308 · Cited by 481
A receding-horizon regulator for nonlinear systems and a neural approximation
Parisini, T., & Zoppoli, R. (1995) · Automatica · doi:10.1016/0005-1098(95)00044-w · Cited by 366
Richter, S., Jones, C. N., & Morari, M. (2011) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2011.2176389 · Cited by 291
Universal Function Approximation by Deep Neural Nets with Bounded Width and ReLU Activations
Hanin, B. (n.d.) · RePEc: Research Papers in Economics · Cited by 278
Bounding and Counting Linear Regions of Deep Neural NetworksPreprint
Serra, T., Tjandraatmadja, C., & Ramalingam, S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1711.02114 · Cited by 271
Learning an Approximate Model Predictive Controller With Guarantees
Hertneck, M., Kohler, J., Trimpe, S., & Allgower, F. (2018) · IEEE Control Systems Letters · doi:10.1109/lcsys.2018.2843682 · Cited by 261
Embedded Online Optimization for Model Predictive Control at Megahertz Rates
Jerez, J. L., Goulart, P. J., Richter, S., Constantinides, G. A., Kerrigan, E. C., & Morari, M. (2014) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2014.2351991 · Cited by 256
Accelerated gradient methods and dual decomposition in distributed model predictive control
Giselsson, P., Doan, M. D., Keviczky, T., Schutter, B. D., & Rantzer, A. (2013) · Automatica · doi:10.1016/j.automatica.2013.01.009 · Cited by 244
Neural-Network-Based Output-Feedback Control Under Round-Robin Scheduling Protocols
Ding, D., Wang, Z., Han, Q., & Wei, G. (2018) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2018.2827037 · Cited by 233
Approximate explicit constrained linear model predictive control via orthogonal search tree
Johansen, T. A., & Grancharova, A. (2003) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2003.811259 · Cited by 215
Learning an approximate model predictive controller with guarantees
Hertneck, M. (2018) · OPUS Publication Server of the University of Stuttgart (University of Stuttgart) · doi:10.18419/opus-9722 · Cited by 210
Approximating Explicit Model Predictive Control Using Constrained Neural Networks
Chen, S., Saulnier, K., Atanasov, N., Lee, D. D., Kumar, V., Pappas, G. J., & Morari, M. (2018) · 2018 Annual American Control Conference (ACC) · doi:10.23919/acc.2018.8431275 · Cited by 209
Statistical Learning Theory: Models, Concepts, and Results
Luxburg, U. V., & Schölkopf, B. (2011) · Handbook of the History of Logic · doi:10.1016/b978-0-444-52936-7.50016-1 · Cited by 189
Liu, C., Gao, J., Li, H., & Xu, D. (2017) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2017.2695499 · Cited by 151
A deep learning-based approach to robust nonlinear model predictive control
Lucia, S., & Karg, B. (2018) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2018.11.038 · Cited by 140
Self-citationDepth-Width Tradeoffs in Approximating Natural Functions with Neural NetworksPreprint
Safran, I., & Shamir, O. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1610.09887 · Cited by 134
DryVR: Data-Driven Verification and Compositional Reasoning for Automotive Systems
Fan, C., Qi, B., Mitra, S., & Viswanathan, M. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-63387-9_22 · Cited by 120
Hierarchical Distributed Model Predictive Control of Interconnected Microgrids
Hans, C. A., Braun, P., Raisch, J., Grüne, L., & Reincke-Collon, C. (2018) · IEEE Transactions on Sustainable Energy · doi:10.1109/tste.2018.2802922 · Cited by 107
Lucia, S., Navarro, D., Lucía, Ó., Zometa, P., & Findeisen, R. (2017) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2017.2719940 · Cited by 105
Self-citationCsekő, L. H., Kvasnica, M., & Lantos, B. (2015) · IEEE Transactions on Control Systems Technology · doi:10.1109/tcst.2014.2382571 · Cited by 98
Predictive control, embedded cyberphysical systems and systems of systems – A perspective
Lucia, S., Kögel, M., Zometa, P., Quevedo, D. E., & Findeisen, R. (2016) · Annual Reviews in Control · doi:10.1016/j.arcontrol.2016.04.002 · Cited by 85
Self-citationμAO-MPC: A free code generation tool for embedded real-time linear model predictive control
Zometa, P., Kögel, M., & Findeisen, R. (2013) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.1109/acc.2013.6580668 · Cited by 59
Deep learning-based embedded mixed-integer model predictive control
Karg, B., & Lucia, S. (2018) · European Control Conference (ECC) · doi:10.23919/ecc.2018.8550234 · Cited by 53
Self-citationOptimized decision trees for point location in polytopic data sets - application to explicit MPC
Fuchs, A., Jones, C. N., & Morari, M. (2010) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.1109/acc.2010.5530979 · Cited by 23
Summers, S., Jones, C. N., Lygeros, J., & Morari, M. (2009) · IEEE Conference on Decision and Control (CDC) · doi:10.1109/cdc.2009.5400583 · Cited by 14
Metadata
- Authors
- Karg, B., & Lucia, S.
- Venue
- IEEE Transactions on Cybernetics
- Published
- 2020-09
- Volume
- 50
- Issue
- 9
- Pages
- 3866-3878
- ISSN
- 2168-2267 (ISSN-L)
- 2168-2275
- OpenAlex
- W2810195925
- Cited by
- 306
- References
- 46
- FWCI
- 17.38
- Type
- journal article
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.