Kybernetika 60 no. 4, 513-534, 2024

A new method based on least-squares support vector regression for solving optimal control problems

Mitra Bolhassani, Hassan Dana Mazraeh and Kourosh ParandDOI: 10.14736/kyb-2024-4-0513

Abstract:

In this paper, a new application of the Least Squares Support Vector Regression (LS-SVR) with Legendre basis functions as mapping functions to a higher dimensional future space is considered for solving optimal control problems. At the final stage of LS-SVR, an optimization problem is formulated and solved using Maple optimization packages. The accuracy of the method are illustrated through numerical examples, including nonlinear optimal control problems. The results demonstrate that the proposed method is capable of solving optimal control problems with high accuracy.

Keywords:

Least squares support vector machines, Optimal control problems, Legendre orthogonal polynomials, Regression, Artificial intelligence

Classification:

68T20, 49Mxx

paper.pdf

References:

  1. H. M. Amman and D. A. Kendrick: Computing the 58 (1998), 2, 185-191.    DOI:10.1016/S0165-1765(97)00263-2
  2. J. T. Betts: Survey of numerical methods for trajectory optimization. J. Guidance Control Dynamics (1998), 98124-2207.   CrossRef
  3. H. G. Bock and K. J. Plitt: A multiple shooting algorithm for direct solution of optimal control problems. In: IFAC 9th congress 1984, pp. 1603-1608.   DOI:10.1016/S1474-6670(17)61205-9
  4. C. Cortes and V. Vapnik: Support-vector networks. Mach. Learn 20 (1995), 3, 273-297.   DOI:10.1007/BF00994018
  5. G. N. Elnagar and M. Razzaghi: A collocation-type method for linear quadratic optimal control problems. Opt. Control Appl. Methods 8 (1997), 3, 227-235.   CrossRef
  6. B. Kafash, A. Delavarkhalafi and S. M. Karbassi: Application of Chebyshev polynomials to derive efficient algorithms for the solution of optimal control problems. Scientia Iranica 19 (2012), 3, 795-805.   DOI:10.1016/j.scient.2011.06.012
  7. S. Kang, J. Wang, C. Li and J. Shan: Nonlinear optimal control with disturbance rejection for asteroid landing. J. Franklin Inst. 355 (2018), 16, 8027-8048.   DOI:10.1016/j.jfranklin.2018.06.028
  8. D. E. Kirk: Optimal Control Theory. Prentice-Hall, Englewood Cliffs 1970.   CrossRef
  9. S. Latifi, K. Parand and M. Delkhosh: Generalized Lagrange-Jacobi-Gauss-Radau collocation method for solving a nonlinear optimal control problem with the classical diffusion equation. European Phys. J. Plus, 2020.   CrossRef
  10. R. C. Loxton, K. L. Teo, V. Rehbock and W. K. Ling: Optimal switching instants for a switched-capacitor DC/DC power converter. Automatica J. IFAC 45 (2009), 4, 973-980.   DOI:10.1016/j.automatica.2008.10.031
  11. H. R. Marzban and M. Razzaghi: Optimal control of linear systems via hybrid of block-pulse functions and Legendre polynomials. J. Franklin Inst. 341 (2004), 279-293.   DOI:10.1016/j.jfranklin.2003.12.011
  12. K. M. Mohammadi, M. Jani and K. Parand: A least squares support vector regression for anisotropic diffusion filtering. arXiv preprint arXiv, 2022.   CrossRef
  13. E. Naevdal: Solving continuous-time optimal-control problems with a spreadsheet. J. Econom. Educ. 34 (2003), 2, 99-122.   DOI:10.3928/0022-0124-20030501-03
  14. L. O. Naraigh and A. Byrne: Piecewise-constant optimal control strategies for controlling the outbreak of COVID-19 in the Irish population. Math. Biosci 330 (2020), 108496.   DOI:10.1016/j.mbs.2020.108496
  15. K. Parand, S. Latifi, M. Delkhosh and M. M. Moayeri: Generalized Lagrangian Jacobi-Gauss-Radau collocation method for solving a nonlinear 2-D optimal control problem with the classical diffusion equation. arXiv preprint arXiv, 2018.   CrossRef
  16. K. Parand, M. Razzaghi, R. Sahleh and M. Jani: Least squares support vector regression for solving Volterra integral equations    CrossRef
  17. A. Pakniyat, K. Parand and M. Jani: Least squares support vector regression for differential equations on unbounded domains. Chaos Solitons Fractals, 2021.   CrossRef
  18. L. S. Pontryagin, V. Boltyanskii, R. Gamkrelidze and E. Mischenko: The Mathematical Theory of Optimal Processes. Wiley Interscience, 1962.   CrossRef
  19. K. Rabiei and K. Parand: Collocation method to solve inequality-constrained optimal control problems of arbitrary order. Engrg. Comput. 36 (2020), 1, 115-125.   CrossRef
  20. M. Razzaghi, J. Nazarzadeh and K. Y. Nikravesh: A collocation method for optimal control of linear systems with inequality constraints. Math.Problems Engrg. 3 (1998), 6, 503-515.   CrossRef
  21. R. Rifkin, G. Yeo and T. Poggio: Regularized least-squares classification. Nato Sci. Ser. Sub Ser. III Comput. Syst. Sci. 190 (2003), 131-154.   CrossRef
  22. A. Schwartz: Theory and Implementation of Numerical Mathod Based on Runge-Kutta Integration for Solving Optimal Control Problems. Ph.D. Thesis, University of California, 1996.   CrossRef
  23. J. Shen, T. Tang and L-L. Wang: Spectral Methods: Algorithms, Analysis and Applications. Springer Science and Business Media, 2011.   CrossRef
  24. J. A. Suykens and J. Vandewalle: Least squares support vector machine classifiers. Neural Process. Lett. 9 (1999), 3, 293-300.   DOI:10.1023/A:1018628609742
  25. J. A. K. Suykens, T. Van Gestel, J. De Brabanter, B. De Moor and J. Vandewalle: Least Squares Support Vector Machines. World Scientific, 2002.   CrossRef
  26. K. L. Teo, C. J. Goh and K. H. Wong: Unified Computational Approach to Optimal Control Problems. Longmann Scientific and Technical, 1991.   CrossRef
  27. S. Sabermahani, Y. Ordokhani, K. Rabiei and M. Razzaghi: Solution of optimal control problems governed by Volterra integral and fractional integro-differential equations. J. Vibration Control 29 (2023), 15.-16, 3796-3808.   DOI:10.1177/10775463221105923
  28. K. Rabiei and M. Razzaghi: An approach to solve fractional optimal control problems via fractional-order Boubaker wavelets. J. Vibration Control 29 (2023), 7-8, 1806-1819.   DOI:10.1177/10775463211070902
  29. M. H. Heydari, M. Razzaghi and Z. Avazzadeh: Orthonormal piecewise Bernoulli functions: Application for optimal control problems generated. Vibration Control 29 (2023), 5-6, 1164-1175.   DOI:10.1177/10775463211059364
  30. M. H. Heydari, R. Tavakoli and M. Razzagh: Application of the extended Chebyshev cardinal wavelets in solving fractional optimal control problems with ABC fractional derivative. Int. J. Systems Sci. 53 (2022), 12, 2694-2708.   DOI:10.1080/00207721.2022.2058640
  31. G. Ghanbari and M. Razzaghi: Numerical solutions for fractional optimal control problems by using generalised fractional-order Chebyshev wavelets. Int. J. Systems Sci. 53 (2022), 4, 778-792.   DOI:10.1080/00207721.2021.1972357
  32. M. H. Heydari and M. Razzaghi: A new class of orthonormal basis functions: application for fractional optimal control problems. Int. J. Systems Sci. 53 (2022), 2, 240-252.   CrossRef
  33. M. Lakestani, Y. Edrisi-Tabrizi and M. Razzaghi: Study of B-spline collocation method for solving fractional optimal control problems. Trans. Inst. Measurement Control 43 (2021), 11, 2425-2437.   DOI:10.1177/0142331220987537
  34. H. Dehestani, Y. Ordokhani and M. Razzaghi: Fractional-order Bessel wavelet functions for solving variable order fractional optimal control problems with estimation error. Int. J. Systems Sci. 51 (2020), 6, 1032-1052.   DOI:10.1080/00207721.2020.1746980
  35. S. Mashayekhi and M. Razzaghi: An approximate method for solving fractional optimal control problems by hybrid functions. J. Vibration Control 24 (2018), 9, 1621-1631.   CrossRef
  36. J. K. Mohammed and A. R. Khudair: A novel numerical method for solving optimal control problems using fourth-degree hat functions. J. Partial Differential Equations Appl. Math. 7 (2023), 9, 100507.   DOI:10.1016/j.padiff.2023.100507
  37. H. Hassani, J. A. Tenreiro Machado, Z. Avazzadeh, E. Naraghirad and M. Sh. Dahaghin: Generalized Bernoulli polynomials: Solving nonlinear 2D fractional optimal control problems. J. Scient. Comput. 83 (2020), 2, 1-21.   CrossRef
  38. J. A. K. Suykens, J. Vandewalle and B. D. Moor: Optimal control by least squares support vector machines. J. Partial Differential Equations Appl. Math. 14 (2001), 1, 23-35.   CrossRef