Topology of the landscape of optimally controlled transitions in a multilevel system

Abstract

According to a theoretical analysis, the control landscape of many quantum control problems has a very favourable topology regardless of the detailed nature of the Hamiltonian, provided that one has full control of the system. It is obvious that full control is not possible in a practical experiment and this study will investigate the influence of experimental and other limitations on the control landscape. Depending on the outcome of the investigation, the applicability of various optimisation techniques will be investigated. In particular, gradient-based optimisation techniques will be investigated and their results will be compared with the results obtained by the more traditional (in the sense of quantum control schemes) genetic type optimisation techniques.