Model predictive control for robust quantum state preparation

A critical engineering challenge in quantum technology is the accurate control of dynamics. Model-based methods for optimal have been shown to be highly effective when theory and experiment closely match. Consequently, realizing high-fidelity processes with model-based requires careful device characterization. In processors based on cold atoms, Hamiltonian can well-characterized. For superconducting qubits operating at milli-Kelvin temperatures, not as Unaccounted physics (i.e., mode discrepancy), coherent disturbances, increased noise compromise traditional control. This work introduces model predictive control (MPC) applications. MPC a closed-loop optimization framework that (i) inherits natural degree disturbance rejection by incorporating measurement feedback, (ii) utilizes finite-horizon optimizations complex multi-input, multi-output dynamical systems under state input constraints, (iii) flexible enough develop synergistically alongside other modern strategies. We show how used generate practical optimized sequences representative examples preparation. Specifically, we demonstrate qubit, weakly-anharmonic system undergoing crosstalk, realize successful even model inadequate. These showcase why an important addition suite.