Attaining optimal controls for manipulating quantum systems

Wusheng Zhu, Herschel Rabitz

Research output: Contribution to journalArticlepeer-review

13 Scopus citations


This article reviews the procedures for attaining laser optimal control fields to manipulate the dynamics of quantum systems. Optimal control theory, whether implemented as a computational technique or directly in the laboratory, is the most general means to achieve precise control of quantum systems. In the case of theoretical control design, a class of monotonically convergent iteration algorithms are presented, which can reduce the computational effort involved. Implementation of optimal control field design can provide physical insight into the feasibility and mechanism of control, even for quantum systems whose Hamiltonian might contain some degree of uncertainty. Moreover, optimal quantum control fields may be directly determined in the laboratory by an analogous procedure with the molecule subject to manipulation acting as an analog computer to guide the discovery of successful controls. To facilitate this process closed-loop iteration is set up with the loop consisting of the laser, the molecular sample, and a learning algorithm to recognize patterns in the control fields giving rise to successful experimental outcomes. Theoretical design and laboratory learning can act in tandem with each other to achieve the best control and reveal information on how the control process occurs.

Original languageEnglish (US)
Pages (from-to)50-58
Number of pages9
JournalInternational Journal of Quantum Chemistry
Issue number2
StatePublished - May 20 2003

All Science Journal Classification (ASJC) codes

  • Atomic and Molecular Physics, and Optics
  • Condensed Matter Physics
  • Physical and Theoretical Chemistry


  • Closed loop
  • Genetic algorithms
  • Monotonically convergent algorithms
  • Open loop
  • Optimal control


Dive into the research topics of 'Attaining optimal controls for manipulating quantum systems'. Together they form a unique fingerprint.

Cite this