论文标题
稳健的量子控制和障碍的进化
Robust quantum control with disorder-dressed evolution
论文作者
论文摘要
最佳量子控制的理论可确定有效产生所需目标状态的时间依赖性控制汉尔顿人。因此,它在量子技术的成功设计和开发中起着至关重要的作用。但是,传递的控制脉冲通常对小扰动非常敏感,这可能会使不可能在实验中可靠地部署这些脉冲。强大的量子控制旨在通过找到控制脉冲来维护其重现目标状态的能力,即使在存在脉冲扰动的情况下,也可以减轻此问题。但是,找到这种健壮的对照脉冲通常很难,因为对照脉冲的评估要求将所有可能的扭曲版本纳入评估。在这里,我们表明,可以根据无序的进化方程来识别稳健的控制脉冲。后者捕获了在描述无序平均密度矩阵演化的量子主方程方面,在这里捕获了脉搏扰动的疾病的影响。在这种稳健控制的方法中,最终状态的纯度表明了基本的控制脉冲的稳健性,如果最终状态是纯净的(并且与目标状态相吻合),则将稳健的控制脉冲单击。我们表明,可以成功地使用该原则来找到强大的控制脉冲。为此,我们适应了克罗托夫(Krotov)的方法,用于疾病的演化,并通过几个单量器控制任务证明了其应用。
The theory of optimal quantum control serves to identify time-dependent control Hamiltonians that efficiently produce desired target states. As such, it plays an essential role in the successful design and development of quantum technologies. However, often the delivered control pulses are exceedingly sensitive to small perturbations, which can make it hard if not impossible to reliably deploy these in experiments. Robust quantum control aims at mitigating this issue by finding control pulses that uphold their capacity to reproduce the target states even in the presence of pulse perturbations. However, finding such robust control pulses is generically hard, since the assessment of control pulses requires the inclusion of all possible distorted versions in the evaluation. Here we show that robust control pulses can be identified based on disorder-dressed evolution equations. The latter capture the effect of disorder, which here stands for the pulse perturbations, in terms of quantum master equations describing the evolution of the disorder-averaged density matrix. In this approach to robust control, the purities of the final states indicate the robustness of the underlying control pulses, and robust control pulses are singled out if the final states are pure (and coincide with the target states). We show that this principle can be successfully employed to find robust control pulses. To this end, we adapt Krotov's method for disorder-dressed evolution and demonstrate its application with several single-qubit control tasks.