Skip to main content
SpringerLink
Log in

On the Sample Complexity of Stabilizing Linear Dynamical Systems from Data

  • Published:
Foundations of Computational Mathematics Aims and scope Submit manuscript

Abstract

Learning controllers from data for stabilizing dynamical systems typically follows a two-step process of first identifying a model and then constructing a controller based on the identified model. However, learning models means identifying generic descriptions of the dynamics of systems, which can require large amounts of data and extracting information that are unnecessary for the specific task of stabilization. The contribution of this work is to show that if a linear dynamical system has dimension (McMillan degree) \(n\), then there always exist \(n\) states from which a stabilizing feedback controller can be constructed, independent of the dimension of the representation of the observed states and the number of inputs. By building on previous work, this finding implies that any linear dynamical system can be stabilized from fewer observed states than the minimal number of states required for learning a model of the dynamics. The theoretical findings are demonstrated with numerical experiments that show the stabilization of the flow behind a cylinder from less data than necessary for learning a model.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

References

  1. Antoulas, A.C.: Approximation of Large-Scale Dynamical Systems, Adv. Des. Control, vol. 6. SIAM, Philadelphia, PA (2005). https://doi.org/10.1137/1.9780898718713

  2. Armstrong, E.: An extension of Bass’ algorithm for stabilizing linear continuous constant systems. IEEE Trans. Autom. Control 20(1), 153–154 (1975). https://doi.org/10.1109/TAC.1975.1100874

    Article  MATH  Google Scholar 

  3. Armstrong, E., Rublein, G.: A stabilization algorithm for linear discrete constant systems. IEEE Trans. Autom. Control 21(4), 629–631 (1976). https://doi.org/10.1109/TAC.1976.1101295

    Article  MATH  Google Scholar 

  4. Beattie, C.A., Gugercin, S.: Realization-independent \(\cal{H}_2\)-approximation. In: 51st IEEE Conference on Decision and Control (CDC), pp. 4953–4958 (2012). https://doi.org/10.1109/CDC.2012.6426344

  5. Behr, M., Benner, P., Heiland, J.: Example setups of Navier-Stokes equations with control and observation: Spatial discretization and representation via linear-quadratic matrix coefficients. e-print arXiv:1707.08711, arXiv (2017). https://doi.org/10.48550/arXiv.1707.08711. Mathematical Software (cs.MS)

  6. Benner, P., Castillo, M., Quintana-Ortí, E.S., Hernández, V.: Parallel partial stabilizing algorithms for large linear control systems. J. Supercomput. 15(2), 193–206 (2000). https://doi.org/10.1023/A:1008108004247

    Article  MATH  Google Scholar 

  7. Benner, P., Gugercin, S., Willcox, K.: A survey of projection-based model reduction methods for parametric dynamical systems. SIAM Rev. 57(4), 483–531 (2015). https://doi.org/10.1137/130932715

    Article  MATH  Google Scholar 

  8. Benner, P., Heiland, J., Werner, S.W.R.: Robust output-feedback stabilization for incompressible flows using low-dimensional \(\cal H\it _{\infty }\)-controllers. Comput. Optim. Appl. 82(1), 225–249 (2022). https://doi.org/10.1007/s10589-022-00359-x

    Article  MATH  Google Scholar 

  9. Benner, P., Li, J.R., Penzl, T.: Numerical solution of large-scale Lyapunov equations, Riccati equations, and linear-quadratic optimal control problems. Numer. Lin. Alg. Appl. 15(9), 755–777 (2008). https://doi.org/10.1002/nla.622

    Article  MATH  Google Scholar 

  10. Benner, P., Schilders, W., Grivet-Talocia, S., Quarteroni, A., Rozza, G., Silveira, L.M.: Model Order Reduction. Volume 1: System- and Data-Driven Methods and Algorithms. De Gruyter, Berlin, Boston (2021). https://doi.org/10.1515/9783110498967

  11. Benner, P., Schilders, W., Grivet-Talocia, S., Quarteroni, A., Rozza, G., Silveira, L.M.: Model Order Reduction. Volume 2: Snapshot-Based Methods and Algorithms. De Gruyter, Berlin, Boston (2021). https://doi.org/10.1515/9783110671490

  12. Benner, P., Werner, S.W.R.: MORLAB – Model Order Reduction LABoratory (version 5.0) (2019). https://doi.org/10.5281/zenodo.3332716. See also: https://www.mpi-magdeburg.mpg.de/projects/morlab

  13. Benner, P., Werner, S.W.R.: MORLAB—The Model Order Reduction LABoratory. In: P. Benner, T. Breiten, H. Faßbender, M. Hinze, T. Stykel, R. Zimmermann (eds.) Model Reduction of Complex Dynamical Systems, International Series of Numerical Mathematics, vol. 171, pp. 393–415. Birkhäuser, Cham (2021). https://doi.org/10.1007/978-3-030-72983-7_19

  14. Berberich, J., Koch, A., Scherer, C.W., Allgöwer, F.: Robust data-driven state-feedback design. In: 2020 American Control Conference (ACC), pp. 1532–1538 (2020). https://doi.org/10.23919/ACC45564.2020.9147320

  15. Boyd, S., Ghaoui, L.E., Feron, E., Balakrishnan, V.: Linear Matrix Inequalities in Systems and Control Theory, Studies in Applied and Numerical Mathematics, vol. 15. SIAM, Philadelphia, PA (1994). https://doi.org/10.1137/1.9781611970777

  16. Breiten, T., Morandin, R., Schulze, P.: Error bounds for port-Hamiltonian model and controller reduction based on system balancing. Comput. Math. Appl. 116, 100–115 (2021). https://doi.org/10.1016/j.camwa.2021.07.022

    Article  MATH  Google Scholar 

  17. Brunton, S.L., Brunton, B.W., Proctor, J.L., Kutz, J.N.: Koopman invariant subspaces and finite linear representations of nonlinear dynamical systems for control. PLoS ONE 11(2), e0150171 (2016). https://doi.org/10.1371/journal.pone.0150171

    Article  Google Scholar 

  18. Brunton, S.L., Kutz, J.N.: Data-driven science and engineering: machine learning, dynamical systems, and control. Cambridge University Press, Cambridge (2019). https://doi.org/10.1017/9781108380690

  19. Brunton, S.L., Proctor, J.L., Kutz, J.N.: Discovering governing equations from data by sparse identification of nonlinear dynamical systems. Proc. Natl. Acad. Sci. U. S. A. 113(15), 3932–3937 (2016). https://doi.org/10.1073/pnas.1517384113

    Article  MATH  Google Scholar 

  20. Campi, M.C., Lecchini, A., Savaresi, S.M.: Virtual reference feedback tuning: a direct method for the design of feedback controllers. Automatica J. IFAC 38(8), 1337–1346 (2002). https://doi.org/10.1016/S0005-1098(02)00032-8

    Article  MATH  Google Scholar 

  21. Datta, B.N.: Numerical Methods for Linear Control Systems: Design and Analysis. Academic Press, San Diego, CA (2004). https://doi.org/10.1016/B978-0-12-203590-6.X5000-9

  22. De Hoop, M.V., Kovachki, N.B., Nelsen, N.H., Stuart, A.M.: Convergence rates for learning linear operators from noisy data. e-print 2108.12515, arXiv (2021). https://doi.org/10.48550/arXiv.2108.12515. Statistics Theory (math.ST)

  23. De Persis, C., Tesi, P.: Formulas for data-driven control: Stabilization, optimality, and robustness. IEEE Trans. Autom. Control 65(3), 909–924 (2020). https://doi.org/10.1109/TAC.2019.2959924

    Article  MATH  Google Scholar 

  24. Dean, S., Mania, H., Matni, N., Recht, B., S., T.: On the sample complexity of the linear quadratic regulator. Found. Comput. Math. 20(4), 633–679 (2020). https://doi.org/10.1007/s10208-019-09426-y

  25. Dragan, V., Halanay, A.: Stabilization of Linear Systems. Systems & Control: Foundations & Applications. Birkhäuser, Boston, MA (1999). https://doi.org/10.1007/978-1-4612-1570-7

  26. Drmač, Z., Mezić, I., Mohr, R.: Data driven modal decompositions: Analysis and enhancements. SIAM J. Sci. Comput. 40(4), A2253–A2285 (2018). https://doi.org/10.1137/17M1144155

    Article  MATH  Google Scholar 

  27. Fliess, M., Join, C.: Model-free control. Int. J. Control 86(12), 2228–2252 (2013). https://doi.org/10.1080/00207179.2013.810345

    Article  MATH  Google Scholar 

  28. Gevers, M., Bazanella, A.S., Bombois, X., Miskovic, L.: Identification and the information matrix: How to get just sufficiently rich? IEEE Trans. Autom. Control 54(12), 2828–2840 (2009). https://doi.org/10.1109/TAC.2009.2034199

    Article  MATH  Google Scholar 

  29. Golub, G.H., Van Loan, C.F.: Matrix Computations, fourth edn. Johns Hopkins Studies in the Mathematical Sciences. Johns Hopkins University Press, Baltimore (2013)

    Book  MATH  Google Scholar 

  30. Gosea, I.V., Gugercin, S., Beattie, C.: Data-driven balancing of linear dynamical systems. SIAM J. Sci. Comput. 44(1), A554–A582 (2022). https://doi.org/10.1137/21M1411081

    Article  MATH  Google Scholar 

  31. Grant, M., Boyd, S.: CVX: Matlab software for disciplined convex programming, version 2.2. http://cvxr.com/cvx (2020)

  32. Grant, M.C., Boyd, S.P.: Graph implementations for nonsmooth convex programs. In: V.D. Blondel, S.P. Boyd, H. Kimura (eds.) Recent Advances in Learning and Control, Lect. Notes Control Inf. Sci., vol. 371, pp. 95–110. Springer, London (2008). https://doi.org/10.1007/978-1-84800-155-8_7

  33. Gurobi Optimization, LLC: Gurobi optimizer reference manual (2022). https://www.gurobi.com

  34. Ho, B.L., Kalman, R.E.: Editorial: Effective construction of linear state-variable models from input/output functions. at-Automatisierungstechnik 14(1–12), 545–548 (1966). https://doi.org/10.1524/auto.1966.14.112.545

  35. Jonckheere, E.A., Silverman, L.M.: A new set of invariants for linear systems–application to reduced order compensator design. IEEE Trans. Autom. Control 28(10), 953–964 (1983). https://doi.org/10.1109/TAC.1983.1103159

    Article  MATH  Google Scholar 

  36. Kaiser, E., Kutz, J.N., Brunton, S.L.: Sparse identification of nonlinear dynamics for model predictive control in the low-data limit. Proc. R. Soc. A: Math. Phys. Eng. Sci. 474(2219), 20180335 (2018). https://doi.org/10.1098/rspa.2018.0335

    Article  MATH  Google Scholar 

  37. Kaiser, E., Kutz, J.N., Brunton, S.L.: Data-driven discovery of Koopman eigenfunctions for control. Mach. Learn.: Sci. Technol. 2(3), 035023 (2021). https://doi.org/10.1088/2632-2153/abf0f5

  38. Kramer, B., Peherstorfer, B., Willcox, K.: Feedback control for systems with uncertain parameters using online-adaptive reduced models. SIAM J. Appl. Dyn. Syst. 16(3), 1563–1586 (2017). https://doi.org/10.1137/16M1088958

    Article  MATH  Google Scholar 

  39. Lancaster, P., Rodman, L.: Algebraic Riccati Equations. Oxford Science Publications. The Clarendon Press, Oxford University Press, New York (1995)

    MATH  Google Scholar 

  40. Leibfritz, F.: \(COMPl_{e}ib\): COnstrained Matrix-optimization Problem library – a collection of test examples for nonlinear semidefinite programs, control system design and related problems. Tech.-report, University of Trier (2004). http://www.friedemann-leibfritz.de/COMPlib_Data/COMPlib_Main_Paper.pdf

  41. Lequin, O., Gevers, M., Mossberg, M., Bosmans, E., Triest, L.: Iterative feedback tuning of PID parameters: comparison with classical tuning rules. Control Eng. Pract. 11(9), 1023–1033 (2003). https://doi.org/10.1016/S0967-0661(02)00303-9

    Article  Google Scholar 

  42. Mayo, A.J., Antoulas, A.C.: A framework for the solution of the generalized realization problem. Linear Algebra Appl. 425(2–3), 634–662 (2007). https://doi.org/10.1016/j.laa.2007.03.008. Special issue in honor of P. A. Fuhrmann, Edited by A. C. Antoulas, U. Helmke, J. Rosenthal, V. Vinnikov, and E. Zerz

  43. MOSEK ApS: The MOSEK optimization toolbox for MATLAB manual. Version 9.1.9 (2019). https://docs.mosek.com/9.1/toolbox/index.html

  44. Peherstorfer, B.: Multifidelity Monte Carlo estimation with adaptive low-fidelity models. SIAM/ASA J. Uncertainty Quantification 7(2), 579–603 (2019). https://doi.org/10.1137/17M1159208

    Article  MATH  Google Scholar 

  45. Peherstorfer, B.: Sampling low-dimensional Markovian dynamics for preasymptotically recovering reduced models from data with operator inference. SIAM J. Sci. Comput. 42(5), A3489–A3515 (2020). https://doi.org/10.1137/19M1292448

    Article  MATH  Google Scholar 

  46. Peherstorfer, B., Gugercin, S., Willcox, K.: Data-driven reduced model construction with time-domain Loewner models. SIAM J. Sci. Comput. 39(5), A2152–A2178 (2017). https://doi.org/10.1137/16M1094750

    Article  MATH  Google Scholar 

  47. Peherstorfer, B., Willcox, K.: Data-driven operator inference for nonintrusive projection-based model reduction. Comput. Methods Appl. Mech. Eng. 306, 196–215 (2016). https://doi.org/10.1016/j.cma.2016.03.025

    Article  MATH  Google Scholar 

  48. Quarteroni, A., Rozza, G.: Reduced Order Methods for Modeling and Computational Reduction, MS &A – Modeling, Simulation and Applications, vol. 9. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-02090-7

  49. Rosenbrock, H.H.: State-space and Multivariable Theory, Studies in dynamical systems, vol. 3. Wiley, New York (1970)

    MATH  Google Scholar 

  50. Safonov, M.G., Tsao, T.C.: The unfalsified control concept: A direct path from experiment to controller. In: B.A. Francis, A.R. Tannenbaum (eds.) Feedback Control, Nonlinear Systems, and Complexity, Lect. Notes Control Inf. Sci., vol. 202, pp. 196–214. Springer, Berlin, Heidelberg (1995). https://doi.org/10.1007/BFb0027678

  51. Schaeffer, H., Caflisch, R., Hauck, C.D., Osher, S.: Sparse dynamics for partial differential equations. Proc. Natl. Acad. Sci. U. S. A. 110(17), 6634–6639 (2013). https://doi.org/10.1073/pnas.1302752110

    Article  MATH  Google Scholar 

  52. Schaeffer, H., Tran, G., Ward, R.: Extracting sparse high-dimensional dynamics from limited data. SIAM J. Appl. Math. 78(6), 3279–3295 (2018). https://doi.org/10.1137/18M116798X

    Article  MATH  Google Scholar 

  53. Schmid, P.J.: Dynamic mode decomposition of numerical and experimental data. J. Fluid Mech. 656, 5–28 (2010). https://doi.org/10.1017/S0022112010001217

    Article  MATH  Google Scholar 

  54. Schulze, P., Unger, B.: Data-driven interpolation of dynamical systems with delay. Syst. Control Lett. 97, 125–131 (2016). https://doi.org/10.1016/j.sysconle.2016.09.007

    Article  MATH  Google Scholar 

  55. Schulze, P., Unger, B., Beattie, C., Gugercin, S.: Data-driven structured realization. Linear Algebra Appl. 537, 250–286 (2018). https://doi.org/10.1016/j.laa.2017.09.030

    Article  MATH  Google Scholar 

  56. Sturm, J.F.: Using SeDuMi 1.02, A Matlab toolbox for optimization over symmetric cones. Optim. Methods Softw. 11(1–4), 625–653 (1999). https://doi.org/10.1080/10556789908805766

  57. Tu, J.H., Rowley, C.W., Luchtenburg, D.M., Brunton, S.L., Kutz, J.N.: On dynamic mode decomposition: Theory and applications. J. Comput. Dyn. 1(2), 391–421 (2014). https://doi.org/10.3934/jcd.2014.1.391

    Article  MATH  Google Scholar 

  58. Tu, S., Boczar, R., Packard, A., Recht, B.: Non-asymptotic analysis of robust control from coarse-grained identification. e-print 1707.04791, arXiv (2017). https://doi.org/10.48550/arXiv.1707.04791. Optimization and Control (math.OC)

  59. Van Overschee, P., De Moor, B.: Subspace Identification for Linear Systems: Theory, Implementation, Applications. Springer, Boston, MA (1996). https://doi.org/10.1007/978-1-4613-0465-4

  60. Van Waarde, H.J., Eising, J., Trentelman, H.L., Camlibel, M.K.: Data informativity: A new perspective on data-driven analysis and control. IEEE Trans. Autom. Control 65(11), 4753–4768 (2020). https://doi.org/10.1109/TAC.2020.2966717

    Article  MATH  Google Scholar 

  61. Voigt, M.: On linear-quadratic optimal control and robustness of differential-algebraic systems. Dissertation, Otto-von-Guericke-Universität, Magdeburg, Germany (2015)

  62. Werner, S.W.R.: Code, data and results for numerical experiments in “On the sample complexity of stabilizing linear dynamical systems from data” (version 1.0) (2022). https://doi.org/10.5281/zenodo.5902997

  63. Willems, J.C., Rapisarda, P., Markovsky, I., De Moor, B.L.M.: A note on persistency of excitation. Syst. Control Lett. 54(4), 325–329 (2005). https://doi.org/10.1016/j.sysconle.2004.09.003

    Article  MATH  Google Scholar 

  64. Ziegler, J., Nichols, N.: Optimum settings for automatic controllers. Trans. ASME 64, 759–768 (1942)

    Google Scholar 

Download references

Acknowledgements

The authors acknowledge support from the Air Force Office of Scientific Research (AFOSR) award FA9550-21-1-0222 (Dr. Fariba Fahroo). The second author additionally acknowledges support from the National Science Foundation under Grant No. 2012250.

Author information

Authors and Affiliations

  1. Courant Institute of Mathematical Sciences, New York University, 251 Mercer St, New York, NY, 10012, USA

    Steffen W. R. Werner & Benjamin Peherstorfer

Authors
  1. Steffen W. R. Werner
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Benjamin Peherstorfer
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Steffen W. R. Werner.

Additional information

Communicated by Rachel Ward.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Werner, S.W.R., Peherstorfer, B. On the Sample Complexity of Stabilizing Linear Dynamical Systems from Data. Found Comput Math (2023). https://doi.org/10.1007/s10208-023-09605-y

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s10208-023-09605-y

Keywords

Mathematics Subject Classification

Search

Navigation