Stochastic Maximum Principle for Square-Integrable Optimal Control of Linear Stochastic Systems

Abstract

The authors give a stochastic maximum principle for square-integrable optimal control of linear stochastic systems. The control domain is not necessarily convex and the cost functional can have a quadratic growth. In particular, they give a stochastic maximum principle for the linear quadratic optimal control problem.

Keywords

Stochastic maximum principle / Optimal control / Linear stochastic system / Square integrability / 60H10 / 93E20

References

[1] Arkin V I, Saksonov M T. Necessary optimality conditions in problems of the control of stochastic differential equations. Dokl. Akad. Nauk SSSR. 1979, 244(1): 11-15

[2] Bensoussan A. Lectures on stochastic control, Nonlinear filtering and stochastic control (Cortona, 1981). 1982, Berlin, New York, Springer-Verlag: 162

[3] Bensoussan A. Stochastic control by functional analysis methods. 1982, Amsterdam, New York, North-Holland Publishing Co.

[4] Bismut J M. Linear quadratic optimal stochastic control with random coefficients. SIAM J. Control Optim.. 1976, 14(3): 419-444

[5] Cadenillas A, Karatzas I. The stochastic maximum principle for linear convex optimal control with random coefficients. SIAM J. Control Optim.. 1995, 33(2): 590-624

[6] Chen X R, Ni G X. Textbook of Mathematical Statistics. 2009(in Chinese)

[7] Gal’čuk L I. On the existence and uniqueness of solutions of stochastic equations with respect to semi-martingales. Teor. Veroyatnost. i Primenen.. 1978, 23(4): 782-795

[8] Haussmann U G. General necessary conditions for optimal control of stochastic systems. Math. Programming Stud.. 1976, 6: 30-48

[9] Ikeda N, Watanabe S. Stochastic differential equations and diffusion processes. 1989, Amsterdam, North-Holland Publishing Co.Kodansha, Ltd., Tokyo

[10] Ji S L, Xue X L. A stochastic maximum principle for linear quadratic problem with nonconvex control domain. Math. Control Relat. Fields. 2019, 9: 495-507

[11] Kushner H J. Necessary conditions for continuous parameter stochastic optimization problems. SIAM J. Control. 1972, 10: 550-565

[12] Li X P. The Foundation of Probability Theory. 2010, Biejing, Higher Education Press(in Chinese)

[13] Liapunov A A. On completely additive vector functions, III, on a. problem of Ju. Č. Neĭman. Problemy Kihernet. 1964, 12: 165-168

[14] Liapunov A A. On completely additive vector functions, IV. Problemy Kibernet. 1964, 12: 169-179

[15] Øksendal B. Stochastic Differential Equations: An Introduction with Applications. 2003, Berlin, Springer-Verlag

[16] Peng S G. A general stochastic maximum principle for optimal control problems. SIAM J. Control Optim.. 1990, 28(4): 966-979

[17] Tang S J. General linear quadratic optimal stochastic control problems with random coefficients: Linear stochastic Hamilton systems and backward stochastic Riccati equations. SIAM J. Control Optim. 2003, 42(1): 53-75

[18] Wei L S. Mathematical Statistics. 2008, Hefei, Press of University of Science and Technology of China(in Chinese)

[19] Yong J M, Zhou X Y. Stochastic controls: Hainiltonian systems and HJB equations. 1999, New York, Springer-Verlag