Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
175 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Wasserstein Distributionally Robust Regret-Optimal Control in the Infinite-Horizon (2312.17376v1)

Published 28 Dec 2023 in eess.SY, cs.SY, and math.OC

Abstract: We investigate the Distributionally Robust Regret-Optimal (DR-RO) control of discrete-time linear dynamical systems with quadratic cost over an infinite horizon. Regret is the difference in cost obtained by a causal controller and a clairvoyant controller with access to future disturbances. We focus on the infinite-horizon framework, which results in stability guarantees. In this DR setting, the probability distribution of the disturbances resides within a Wasserstein-2 ambiguity set centered at a specified nominal distribution. Our objective is to identify a control policy that minimizes the worst-case expected regret over an infinite horizon, considering all potential disturbance distributions within the ambiguity set. In contrast to prior works, which assume time-independent disturbances, we relax this constraint to allow for time-correlated disturbances, thus actual distributional robustness. While we show that the resulting optimal controller is non-rational and lacks a finite-dimensional state-space realization, we demonstrate that it can still be uniquely characterized by a finite dimensional parameter. Exploiting this fact, we introduce an efficient numerical method to compute the controller in the frequency domain using fixed-point iterations. This method circumvents the computational bottleneck associated with the finite-horizon problem, where the semi-definite programming (SDP) solution dimension scales with the time horizon. Numerical experiments demonstrate the effectiveness and performance of our framework.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (48)
  1. Wasserstein Tube MPC with Exact Uncertainty Propagation, April 2023a. URL http://arxiv.org/abs/2304.12093. arXiv:2304.12093 [math].
  2. Capture, Propagate, and Control Distributional Uncertainty, April 2023b. URL http://arxiv.org/abs/2304.02235. arXiv:2304.02235 [math].
  3. Dimitri Bertsekas. Nonlinear programming, volume 4. Athena scientific, 2016.
  4. Andrea Braides. Gamma-Convergence for Beginners. Oxford University Press, July 2002. ISBN 978-0-19-850784-0. 10.1093/acprof:oso/9780198507840.001.0001. URL https://academic.oup.com/book/1987.
  5. Distributionally robust infinite-horizon control: from a pool of samples to the design of dependable controllers, 2023.
  6. Gianni Dal Maso. An Introduction to Γnormal-Γ\Gammaroman_Γ-Convergence. Birkhäuser Boston, Boston, MA, 1993. ISBN 978-1-4612-6709-6 978-1-4612-0327-8. 10.1007/978-1-4612-0327-8. URL http://link.springer.com/10.1007/978-1-4612-0327-8.
  7. A system-level approach to regret optimal control. IEEE Control Systems Letters, 6:2792–2797, 2022.
  8. State-space solutions to standard h2subscriptℎ2h_{2}italic_h start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT and h∞subscriptℎh_{\infty}italic_h start_POSTSUBSCRIPT ∞ end_POSTSUBSCRIPT control problems. In 1988 American Control Conference, pages 1691–1696, Atlanta, GA, USA, June 1988. IEEE. 10.23919/ACC.1988.4789992. URL https://ieeexplore.ieee.org/document/4789992/.
  9. John C. Doyle. Structured uncertainty in control system design. In 1985 24th IEEE Conference on Decision and Control, pages 260–265, 1985. 10.1109/CDC.1985.268842.
  10. Lasha Ephremidze. An Elementary Proof of the Polynomial Matrix Spectral Factorization Theorem, November 2010. URL http://arxiv.org/abs/1011.3777. arXiv:1011.3777 [math].
  11. On the Algorithmization of Janashia-Lagvilava Matrix Spectral Factorization Method. IEEE Transactions on Information Theory, 64(2):728–737, February 2018. ISSN 0018-9448, 1557-9654. 10.1109/TIT.2017.2772877. URL http://ieeexplore.ieee.org/document/8105834/.
  12. Fatou’s Lemma in Its Classic Form and Lebesgue’s Convergence Theorems for Varying Measures with Applications to MDPs, June 2019. URL http://arxiv.org/abs/1902.01525. arXiv:1902.01525 [math].
  13. Distributionally Robust Stochastic Optimization with Wasserstein Distance, April 2022. URL http://arxiv.org/abs/1604.02199. arXiv:1604.02199 [math].
  14. Asymptotic eigenvalue distribution of block Toeplitz matrices and application to blind SIMO channel identification. IEEE Transactions on Information Theory, 47(3):1243–1251, March 2001. ISSN 00189448. 10.1109/18.915697. URL http://ieeexplore.ieee.org/document/915697/.
  15. Regret-optimal measurement-feedback control. In Learning for Dynamics and Control, pages 1270–1280. PMLR, 2021a.
  16. Regret-optimal control in dynamic environments, February 2021b. URL http://arxiv.org/abs/2010.10473. arXiv:2010.10473 [cs, eess, math].
  17. Regret-optimal estimation and control. IEEE Transactions on Automatic Control, 68(5):3041–3053, 2023.
  18. Robert M. Gray. Toeplitz and Circulant Matrices: A Review. Foundations and Trends® in Communications and Information Theory, 2(3):155–239, 2005. ISSN 1567-2190, 1567-2328. 10.1561/0100000006. URL http://www.nowpublishers.com/article/Details/CIT-006.
  19. Asymptotically Equivalent Sequences of Matrices and Hermitian Block Toeplitz Matrices With Continuous Symbols: Applications to MIMO Systems. IEEE Transactions on Information Theory, 54(12):5671–5680, December 2008. ISSN 0018-9448. 10.1109/TIT.2008.2006401. URL http://ieeexplore.ieee.org/document/4675730/.
  20. Jesú Gutiérrez-Gutiérrez. Block Toeplitz Matrices: Asymptotic Results and Applications. Foundations and Trends® in Communications and Information Theory, 8(3):179–257, 2011. ISSN 1567-2190, 1567-2328. 10.1561/0100000066. URL http://www.nowpublishers.com/article/Details/CIT-066.
  21. Wasserstein Distributionally Robust Regret-Optimal Control under Partial Observability, July 2023a. URL http://arxiv.org/abs/2307.04966. arXiv:2307.04966 [math].
  22. Regret-Optimal Control under Partial Observability, November 2023b. URL http://arxiv.org/abs/2311.06433. arXiv:2311.06433 [cs, eess, math].
  23. Wasserstein distributionally robust control of partially observable linear systems: Tractable approximation and performance guarantee. In 2022 IEEE 61st Conference on Decision and Control (CDC), pages 4800–4807. IEEE, 2022.
  24. Indefinite-Quadratic Estimation and Control. Society for Industrial and Applied Mathematics, 1999. 10.1137/1.9781611970760. URL https://epubs.siam.org/doi/abs/10.1137/1.9781611970760.
  25. Linear estimation. Prentice Hall information and system sciences series. Prentice Hall, Upper Saddle River, N.J, 2000. ISBN 978-0-13-022464-4.
  26. R. E. Kalman. A New Approach to Linear Filtering and Prediction Problems. Journal of Basic Engineering, 82(1):35–45, March 1960. ISSN 0021-9223. 10.1115/1.3662552.
  27. Distributional robustness in minimax linear quadratic control with Wasserstein distance, February 2021. URL http://arxiv.org/abs/2102.12715. arXiv:2102.12715 [cs, eess, math].
  28. F. Leibfritz and W. Lipinski. Description of the benchmark examples in compleib 1.0. Dept. Math, Univ. Trier, Germany, 32, 2003.
  29. Data-Driven Distributionally Robust Optimal Control with State-Dependent Noise, August 2023. URL http://arxiv.org/abs/2303.02293. arXiv:2303.02293 [cs].
  30. Safe control with minimal regret. In Learning for Dynamics and Control Conference, pages 726–738. PMLR, 2022.
  31. Closing the Gap to Quadratic Invariance: a Regret Minimization Approach to Optimal Distributed Control, November 2023. URL http://arxiv.org/abs/2311.02068. arXiv:2311.02068 [cs, eess].
  32. Data-driven distributionally robust optimization using the Wasserstein metric: performance guarantees and tractable reformulations. Mathematical Programming, 171(1-2):115–166, September 2018. ISSN 0025-5610, 1436-4646. 10.1007/s10107-017-1172-1. URL http://link.springer.com/10.1007/s10107-017-1172-1.
  33. H.I. Nurdin. A New Approach to Spectral Factorization of a Class of Matrix-Valued Spectral Densities. In Proceedings of the 44th IEEE Conference on Decision and Control, pages 5929–5934, Seville, Spain, 2005. IEEE. ISBN 978-0-7803-9567-1. 10.1109/CDC.2005.1583110. URL http://ieeexplore.ieee.org/document/1583110/.
  34. C. Rino. Factorization of spectra by discrete Fourier transforms (Corresp.). IEEE Transactions on Information Theory, 16(4):484–485, July 1970. ISSN 0018-9448. 10.1109/TIT.1970.1054502. URL http://ieeexplore.ieee.org/document/1054502/.
  35. Regret-Optimal Filtering for Prediction and Estimation. IEEE Transactions on Signal Processing, 70:5012–5024, 2022. ISSN 1053-587X, 1941-0476. 10.1109/TSP.2022.3212153. URL https://ieeexplore.ieee.org/document/9911672/.
  36. Regret-optimal controller for the full-information problem. In 2021 American Control Conference (ACC), pages 4777–4782. IEEE, 2021.
  37. Data-driven distributionally robust control of energy storage to manage wind power fluctuations. In 2017 IEEE Conference on Control Technology and Applications (CCTA), pages 199–204, 2017. 10.1109/CCTA.2017.8062463.
  38. Filippo Santambrogio. Optimal transport for applied mathematicians. 2015.
  39. Distributionally robust linear quadratic control. arXiv preprint arXiv:2305.17037, 2023.
  40. Dynamic Programming Subject to Total Variation Distance Ambiguity, February 2014. URL http://arxiv.org/abs/1402.1009. arXiv:1402.1009 [math].
  41. Robust Linear Quadratic Regulator for uncertain systems. In 2016 IEEE 55th Conference on Decision and Control (CDC), pages 1515–1520, Las Vegas, NV, USA, December 2016. IEEE. ISBN 978-1-5090-1837-6. 10.1109/CDC.2016.7798481. URL http://ieeexplore.ieee.org/document/7798481/.
  42. P. M. E. M. van der Grinten. Uncertainty in measurement and control. Statistica Neerlandica, 22(1):43–63, 1968. https://doi.org/10.1111/j.1467-9574.1960.tb00617.x.
  43. Cédric Villani. Optimal transport: old and new. Number 338 in Grundlehren der mathematischen Wissenschaften. Springer, Berlin, 2009. ISBN 978-3-540-71049-3. OCLC: ocn244421231.
  44. A distributionally robust approach to regret optimal control using the wasserstein distance, 2023.
  45. Insoon Yang. Wasserstein distributionally robust stochastic control: A data-driven approach. IEEE Transactions on Automatic Control, 66(8):3863–3870, 2020.
  46. G. Zames. Feedback and optimal sensitivity: Model reference transformations, multiplicative seminorms, and approximate inverses. IEEE Transactions on Automatic Control, 26(2):301–320, April 1981. ISSN 0018-9286. 10.1109/TAC.1981.1102603. URL http://ieeexplore.ieee.org/document/1102603/.
  47. Nonlinear wasserstein distributionally robust optimal control. arXiv preprint arXiv:2304.07415, 2023.
  48. Robust and optimal control. Prentice Hall, Upper Saddle River, N.J, 1996. ISBN 978-0-13-456567-5.
Citations (5)

Summary

We haven't generated a summary for this paper yet.