Atomic resolution imaging using quantum sensors enabled by smart adaptive control

这是一个物理学博士研究项目。它的目标是帮助实现实用的原子分辨率磁成像，为生物医学技术和整个社会带来巨大的未来潜力-从解析蛋白质结构到监测诊断中的纳米颗粒。NV中心是金刚石晶体中的氮空位缺陷，具有易于操纵和测量的自旋。它甚至可以耦合到单个原子核的微小磁场，使我们能够检测到它们。要将其开发成一个实用的传感平台，需要克服两个挑战：第一个挑战在于周围晶体中不可避免的缺陷会干扰信号。到目前为止，我们缺乏一个计算上可行的物理上合理的模型来描述这种相互作用。出现了一种新的方法，称为环境的自动压缩，可以准确地捕捉这些影响。第二个挑战是从传感器获得阅读基本上是一个统计过程，需要多次重复测量。这将通过自适应实验控制来缓解。由于在运行中同时优化多个参数在计算上仍然很困难，该项目将使用机器学习的力量来帮助真实的实时识别最佳参数（基于当前的估计/知识状态）。

This is a PhD research project in Physics. Its ambition is to help make practical atomic-resolution magnetic imaging a reality, with significant future potential for biomedical technology and on society at large - from resolving the structure of proteins to monitoring nanoparticles in diagnostics. The NV center is a nitrogen-vacancy defect in a diamond crystal with a spin that is straightforward to manipulate and measure. It couples to even the tiny magnetic fields of individual atomic nuclei, allowing us to detect them. Two challenges need to be overcome to develop this towards a practical platform for sensing: the first lies in the inevitable imperfections in the surrounding crystal that interfere with the signal. So far, we lack a computationally viable yet physically sound model for this interaction. A novel approach has appeared called the automated compression of environments with which those effects can be captured accurately. The second challenge is that getting a reading from the sensor is fundamentally a statistical process and many repeated measurements are required. This will be mitigated through adaptive experiment control. As simultaneously optimizing multiple parameters on the fly remains computationally difficult, this project will use the power of machine learning to aid with the task of identifying optimal parameters (based on the current estimates / state of knowledge) in real time.