Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
139 tokens/sec
GPT-4o
47 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Robust stabilization of polytopic systems via fast and reliable neural network-based approximations (2204.13209v2)

Published 27 Apr 2022 in eess.SY, cs.LG, cs.SY, and math.OC

Abstract: We consider the design of fast and reliable neural network (NN)-based approximations of traditional stabilizing controllers for linear systems with polytopic uncertainty, including control laws with variable structure and those based on a (minimal) selection policy. Building upon recent approaches for the design of reliable control surrogates with guaranteed structural properties, we develop a systematic procedure to certify the closed-loop stability and performance of a linear uncertain system when a trained rectified linear unit (ReLU)-based approximation replaces such traditional controllers. First, we provide a sufficient condition, which involves the worst-case approximation error between ReLU-based and traditional controller-based state-to-input mappings, ensuring that the system is ultimately bounded within a set with adjustable size and convergence rate. Then, we develop an offline, mixed-integer optimization-based method that allows us to compute that quantity exactly.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (56)
  1. Understanding deep neural networks with rectified linear units. In 6th International Conference on Learning Representations, ICLR 2018. OpenReview.net, 2018.
  2. N. Athanasopoulos and G. Bitsoris. Invariant set computation for constrained uncertain discrete-time linear systems. In 49th IEEE Conference on Decision and Control (CDC), pages 5227–5232. IEEE, 2010.
  3. J.-P. Aubin and A. Cellina. Differential inclusions: Set-valued maps and viability theory, volume 264. Springer Science & Business Media, 2012.
  4. A. Bemporad and M. Morari. Control of systems integrating logic, dynamics, and constraints. Automatica, 35(3):407–427, 1999.
  5. The explicit linear quadratic regulator for constrained systems. Automatica, 38(1):3–20, 2002.
  6. F. Blanchini. Ultimate boundedness control for uncertain discrete-time systems via set-induced Lyapunov functions. IEEE Transactions on Automatic Control, 39(2):428–433, 1994.
  7. F. Blanchini. Nonquadratic Lyapunov functions for robust control. Automatica, 31(3):451–461, 1995.
  8. Constrained stabilization with an assigned initial condition set. International Journal of Control, 62(3):601–617, 1995.
  9. F. Blanchini and S. Miani. Set-theoretic methods in control. Birkhäuser, 2015.
  10. On efficient computation of low-complexity controlled invariant sets for uncertain linear systems. International Journal of Control, 83(7):1339–1346, 2010.
  11. Predictive control for linear and hybrid systems. Cambridge University Press, 2017.
  12. A unified view of piecewise linear neural network verification. In Advances in Neural Information Processing Systems, volume 31, pages 4795–4804, 2018.
  13. A simplified neural network for linear matrix inequality problems. Neural Processing Letters, 29(3):213–230, 2009.
  14. Canonical piecewise-linear representation. IEEE Transactions on Circuits and Systems, 35(1):101–111, 1988.
  15. G. Cybenko. Approximation by superpositions of a sigmoidal function. Mathematics of Control, Signals and Systems, 2(4):303–314, 1989.
  16. Triangulations: Structures for algorithms and applications, volume 25. Springer Science & Business Media, 2010.
  17. F. Fabiani and P. J. Goulart. Reliably-stabilizing piecewise-affine neural network controllers. IEEE Transactions on Automatic Control, 68(9):5201–5215, 2023.
  18. Safety verification and robustness analysis of neural networks via quadratic constraints and semidefinite programming. IEEE Transactions on Automatic Control, 67(1):1–15, 2022.
  19. M. Fischetti and J. Jo. Deep neural networks and mixed integer linear optimization. Constraints, 23(3):296–309, 2018.
  20. R. Fletcher. Practical methods of optimization. John Wiley & Sons, 2013.
  21. Deep Learning. MIT Press, 2016.
  22. Piecewise affine functions and polyhedral sets. Optimization, 31(3):209–221, 1994.
  23. D. Guo and Y. Zhang. Zhang neural network for online solution of time-varying linear matrix inequality aided with an equality conversion. IEEE Transactions on Neural Networks and Learning Systems, 25(2):370–382, 2013.
  24. Gurobi Optimization, LLC. Gurobi Optimizer Reference Manual, 2021.
  25. P.-O. Gutman and M. Cwikel. Admissible sets and feedback control for discrete-time linear dynamical systems with bounded controls and states. IEEE Transactions on Automatic Control, 31(4):373–376, 1986.
  26. Neural network design. PWS Publishing Co., 1997.
  27. M. Hausknecht and P. Stone. Deep reinforcement learning in parameterized action space. In Proceedings of the International Conference on Learning Representations (ICLR), 2016.
  28. Neural network adaptive control for a class of nonlinear uncertain dynamical systems with asymptotic stability guarantees. IEEE Transactions on Neural Networks, 19(1):80–89, 2008.
  29. Equivalence of hybrid dynamical models. Automatica, 37(7):1085–1091, 2001.
  30. Multi-Parametric Toolbox 3.0. In 2013 European Control Conference (ECC), pages 502–510. IEEE, 2013.
  31. Multilayer feedforward networks are universal approximators. Neural Networks, 2(5):359–366, 1989.
  32. M. Jordan and A. G. Dimakis. Exactly computing the local Lipschitz constant of ReLU networks. In Advances in Neural Information Processing Systems, volume 33, pages 7344–7353, 2020.
  33. C. Kahlert and L. O. Chua. The complete canonical piecewise-linear representation – Part I: The geometry of the domain space. IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications, 39(3):222–236, 1992.
  34. B. Karg and S. Lucia. Stability and feasibility of neural network-based controllers via output range analysis. In 2020 59th IEEE Conference on Decision and Control (CDC), pages 4947–4954. IEEE, 2020.
  35. B. Karg and S. Lucia. Guaranteed safe control of systems with parametric uncertainties via neural network controllers. In 2022 IEEE 61st Conference on Decision and Control (CDC), pages 7302–7308. IEEE, 2022.
  36. X. Le and J. Wang. Robust pole assignment for synthesizing feedback control systems using recurrent neural networks. IEEE Transactions on Neural Networks and Learning Systems, 25(2):383–393, 2013.
  37. Stabilization of affine systems with polytopic control value sets. Journal of Dynamical and Control Systems, pages 1–13, 2023.
  38. D. Liu and Y. Huang. Neural-network-based online optimal control for uncertain non-linear continuous-time systems with control constraints. IET Control Theory & Applications, 7(17):2037–2047, 2013.
  39. L. Markolf and O. Stursberg. Polytopic input constraints in learning-based optimal control using neural networks. In 2021 European Control Conference (ECC), pages 1018–1023. IEEE, 2021.
  40. Differentiable abstract interpretation for provably robust neural networks. In International Conference on Machine Learning, pages 3578–3586. PMLR, 2018.
  41. Sharp bounds for the number of regions of maxout networks and vertices of Minkowski sums. SIAM Journal on Applied Algebra and Geometry, 6(4):618–649, 2022.
  42. On the number of linear regions of deep neural networks. Advances in Neural Information Processing Systems, 27:2924–2932, 2014.
  43. R. Murray-Smith and T. A. Johansen. Multiple model approaches to modelling and control. Taylor & Francis, 1997.
  44. The role of over-parametrization in generalization of neural networks. In 7th International Conference on Learning Representations, ICLR 2019. OpenReview.net, 2019.
  45. H.-N. Nguyen. Constrained control for uncertain systems: An interpolation based control approach. PhD thesis, Supélec, 2012.
  46. Implicit improved vertex control for uncertain, time-varying linear discrete-time systems with state and control constraints. Automatica, 49(9):2754–2759, 2013.
  47. Explicit improved vertex control for uncertain, time-varying linear discrete-time systems with state and control constraints. International Journal of Robust and Nonlinear Control, 26(12):2652–2667, 2016.
  48. P. Petersen and F. Voigtlaender. Optimal approximation of piecewise smooth functions using deep ReLU neural networks. Neural Networks, 108:296–330, 2018.
  49. The efficient computation of polyhedral invariant sets for linear systems with polytopic uncertainty. In Proceedings of the 2005, American Control Conference, 2005., pages 804–809. IEEE, 2005.
  50. Variational analysis, volume 317. Springer Science & Business Media, 2009.
  51. T. Schindler and A. Dietz. Real-time inference of neural networks on FPGAs for motor control applications. In 2020 10th International Electric Drives Production Conference (EDPC), pages 1–6. IEEE, 2020.
  52. M. Sznaier and M. J. Damborg. An analog “neural net” based suboptimal controller for constrained discrete-time linear systems. Automatica, 28(1):139–144, 1992.
  53. J. Wang and G. Wu. A multilayer recurrent neural network for on-line synthesis of minimum-norm linear feedback control systems via pole assignment. Automatica, 32(3):435–442, 1996.
  54. E. Wong and Z. Kolter. Provable defenses against adversarial examples via the convex outer adversarial polytope. In International Conference on Machine Learning, pages 5286–5295. PMLR, 2018.
  55. Real-time power system state estimation and forecasting via deep unrolled neural networks. IEEE Transactions on Signal Processing, 67(15):4069–4077, 2019.
  56. Near-optimal rapid MPC using neural networks: A primal-dual policy learning framework. IEEE Transactions on Control Systems Technology, 29(5):2102–2114, 2021.
Citations (4)
View on Semantic Scholar

Summary

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

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now