DescriptionAccurate control of complex quantum systems is of great importance for the development of
quantum technologies, as it permits to achieve many goals with high accuracy despite inherent
system imperfections. Realising this in practice, however, is a great challenge, since it requires
precise models and numerically expensive simulations.
The central goal of this project is to develop and implement control techniques that do not
require theoretical modelling, simulation or any knowledge of a systems' microscopic
decomposition. Instead, all necessary information will be obtained directly from the experiment.
We will identify control targets that characterise desired properties of quantum systems well,
and that can be estimated accurately and efficiently in an experiment. Based on the
assessment of these targets and their dependence on tunable control parameters, we will
develop control algorithms such that an optimal control protocol is found within a minimal
number of experimental measurements.
These methods will be developed in direct interplay between simulations of experiments with
many--body systems and actual experimental implementations. In simulations we will target the
creation and stabilisation of many--body localised states and time--crystalline structures, that
will give evidence that the novel control techniques can cope with state--of--the--art quantum
many--body problems. Experimentally we will consider the preparation of highly non--classical
states of a levitated nano--sphere and the formation of large crystals of Rydberg atoms. With
an experiment on an extremely massive quantum object and an experiment with many,
strongly interacting quantum systems, we will be able to experimentally achieve goals that are
clearly out of reach with existing control techniques.
Having verified the efficacy of the control techniques, we will develop a software package and
make it publicly available such that it finds broad application in the development of quantum