Krotov Method for Optimal Control in Closed Quantum Systems

TL;DR

This review paper discusses the Krotov method for optimal control in closed quantum systems, covering mathematical foundations, modifications, and applications in quantum dynamics, comparing it with other control techniques.

Contribution

It provides a comprehensive overview of the Krotov method's mathematical aspects, modifications, and applications in quantum control, highlighting its advantages and differences from other methods.

Findings

Krotov method effectively controls molecular dynamics.

It is suitable for manipulating Bose-Einstein condensates.

The method compares favorably with GRAPE and other control techniques.

Abstract

Mathematical problems of optimal control in quantum systems attract high interest in connection with fundamental questions and existing and prospective applications. An important problem is the development of methods for constructing controls for quantum systems. One of the commonly used methods is the Krotov method initially proposed beyond quantum control in the articles by V.F.~Krotov and I.N.~Feldman (1978, 1983). The method was used to develop a novel approach for finding optimal controls for quantum systems in [D.J. Tannor, V. Kazakov, V. Orlov, In: Time-Dependent Quantum Molecular Dynamics, Boston, Springer, 347--360 (1992)] and [J.~Soml\'{o}i, V.A.~Kazakov, D.J.~Tannor, Chem. Phys., 172:1, 85--98 (1993)], and in many works of various scientists, as described in details in this review. The review discusses mathematical aspects of this method for optimal control of closed quantum systems. It outlines various modifications with respect to defining the improvement function (which in most cases is linear or linear-quadratic), constraints on control spectrum and on the states of a quantum system, regularizers, etc. The review describes applications of the Krotov method to control of molecular dynamics, manipulation of Bose-Einstein condensate, quantum gate generation. We also discuss comparison with GRAPE (GRadient Ascent Pulse Engineering), CRAB (Chopped Random-Basis), the Zhu---Rabitz and the Maday---Turinici methods.