The codifferential descent method in the problem of finding the global minimum of a piecewise affine objective functional in linear control systems

A. V. Fominyh   ; V. V. Karelin   ; L. N. Polyakova   ; S. K. Myshkov   ; V. P. Tregubov

Vestnik Sankt-Peterburgskogo universiteta. Prikladnaâ matematika, informatika, processy upravleniâ, Tome 17 (2021) no. 1, pp. 47-58

Résumé

The article considers the problem of optimal control of an object described by a linear nonstationary system and with a piecewise affine quality functional. The problem is examined in Mayer's form with both free and partially fixed right endpoints. Piecewise continuous and bounded controls that lie in some parallelepiped at each moment of time are admissible. The standard discretization of the original system and the control parametrization are used, some convergence theorems of the discrete problem solution to the continuous problem solution are presented. Further, for the obtained discrete system, the necessary and sufficient minimum conditions are written out in terms of the codifferential, the method of the modified codifferential descent is applied to it, which guarantees to find the global minimum of this problem in a finite number of steps. The proposed algorithm is illustrated with examples.

Détail

Keywords: nonsmooth optimal control problem, piecewise affine function, codifferential, parametrization of control, method of codifferential descent method.

Bibliographie

[1] Fominyh A. V., “Open-loop control of a plant described by a system with nonsmooth right-hand side”, Computational Mathematics and Mathematical Physics, 59:10 (2019), 1639–1648 | DOI | MR | Zbl

[2] Frankowska H., “The first order necessary conditions for nonsmooth variational and control problems”, SIAM J. Control Optim., 22:1 (1984), 1–12 | DOI | MR | Zbl

[3] Ioffe A. D., “Necessary conditions in nonsmooth optimization”, Mathematics of Operations Research, 9:2 (1984), 159–189 | DOI | MR | Zbl

[4] Mordukhovich B., “Necessary conditions for optimality in nonsmooth control problems with nonfixed time”, Differential Equations, 25:1 (1989), 290–299 | MR | Zbl

[5] Shvartsman I. A., “New approximation method in the proof of the maximum principle for nonsmooth optimal control problems with state constraints”, Journal of Mathematical Analysis and Applications, 326:2 (2007), 974–1000 | DOI | MR | Zbl

[6] Vinter R. B., Cheng H., “Necessary conditions for optimal control problems with state constraints”, Transactions of the American Mathematical Society, 350:3 (1998), 1181–1204 | DOI | MR | Zbl

[7] Vinter R. B., “Minimax optimal control”, SIAM J. Control Optim., 44:3 (2005), 939–968 | DOI | MR | Zbl

[8] Gorelik V. A., Tarakanov A. F., “Penalty method and maximum principle for nonsmooth variable-structure control problems”, Cybernetics and Systems Analysis, 28:3 (1992), 432–437 | DOI | MR | Zbl

[9] Gorelik V. A., Tarakanov A. F., “Penalty method for nonsmooth minimax control problems with interdependent variables”, Cybernetics, 25:4 (1989), 483–488 | DOI | MR | Zbl

[10] Morzhin O. V., “On approximation of the subdifferential of the nonsmooth penalty functional in the problems of optimal control”, Automation and Telemechanics, 2009, no. 5, 24–34 (in Russian) | MR | Zbl

[11] Demyanov V. F., Nikulina V. N., Shablinskaya I. R., “Quasidifferentiable functions in optimal control”, Mathematical Programming Study, 29 (1986), 160–175 | DOI | MR | Zbl

[12] Demyanov V. F., Rubinov A. M., Foundations of nonsmooth analysis and quasidifferential calculus, Nauka Publ., M., 1990, 432 pp. (in Russian)

[13] Demyanov V. F., Tamasyan G. Sh., “Direct methods in the parametric moving boundary variational problem”, Numerical Functional Analysis and Optimization, 35:7–9 (2014), 932–961 | DOI | MR | Zbl

[14] Fominyh A. V., Karelin V. V., Polyakova L. N., “Application of the hypodifferential descent method to the problem of constructing an optimal control”, Optimization Letters, 12:8 (2018), 1825–1839 | DOI | MR | Zbl

[15] Fominyh A. V., “Methods of subdifferential and hypodifferential descent in the problem of constructing an integrally constrained program control”, Automation and Remote Control, 78:4 (2017), 608–617 | DOI | MR

[16] Polyakova L. N., “On global unconstrained minimization of the difference of polyhedral functions”, Journal of Global Optimization, 50 (2011), 179–195 | DOI | MR | Zbl

[17] Demyanov V. F., Bagirov A. M., Rubinov A. M., “A method of truncated codifferential with applications to some problems of cluster analysis”, Journal of Global Optimization, 23 (2002), 63–80 | DOI | MR | Zbl

[18] Dolgopolik M. V., “The method of codifferential descent for convex and global piecewise affine optimization”, Optimization Methods and Software, 35:6 (2020), 1191–1222 | DOI | MR | Zbl

[19] Loxton R. C., Teo K. L., Rehbocka V., Yiu K. F. C., “Optimal control problems with a continuous inequality constraint on the state and the control”, Automatica, 45 (2009), 2250–2257 | DOI | MR | Zbl

[20] Teo K. L., Goh C. J., Wong K. H., A unified computational approach to optimal control problems, Longman Scientific and Technical, New York, 1991, 356 pp. | MR | Zbl

[21] Vasil'ev F. P., Optimization methods, Factorial Publ, M., 2002, 824 pp. (in Russian)

[22] Filippov A. F., “On certain questions in the theory of optimal control”, J. SIAM Control. Ser. A, 1 (1962), 76–84 | MR | Zbl

[23] Gorokhovik V. V., Zorko O. I., “Piecewise affine functions and polyhedral sets”, Optimization, 31 (1994), 209–221 | DOI | MR | Zbl

[24] Dolgopolik M. V., “A unifying theory of exactness of linear penalty functions”, Optimization, 65:6 (2015), 1167–1202 | DOI | MR

[25] Dolgopolik M. V., “Exact penalty functions for optimal control problems. II. Exact penalization of terminal and pointwise state constraints”, Optimal Control Applications and Methods, 41:3 (2020), 898–947 | DOI | MR | Zbl

[26] Alekseev V. M., Tikhomirov V. M., Fomin S. V., Optimal control, Nauka Publ, M., 1979, 432 pp. (in Russian) | MR