Non-iterative optimal design of quantum controls
Department of Chemistry, Princeton University, Princeton, NJ 08544, USA
Corresponding author: a firstname.lastname@example.org
Revised: 22 November 2000
Published online: 15 May 2001
A non-iterative means for quantum control design is introduced with the aim of offering practical designs that can later be fine-tuned with laboratory closed-loop techniques. The procedure recognizes that Hamiltonians for realistic system control applications are rarely known accurately. The algorithm takes advantage of this fact by allowing for managed deviations in the equations of motion, thus removing the standard Lagrange multiplier. Suitable time-dependent cost functional weights are introduced that eliminate the traditional final time matching condition, thereby producing non-iterative design equations as an initial value problem. Removal of the final time condition also eliminates the demand that the target state be reached at any artificially imposed time. Tests on a simple molecular system indicate that the algorithm leads to well-behaved designs and that the weight functions are adequately estimated by order of magnitude analysis.
PACS: 02.30.Xx – Calculus of variations / 02.30.Yy – Control theory / 33.80.Wz – Other multiphoton processes
© EDP Sciences, Società Italiana di Fisica, Springer-Verlag, 2001