权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Teaching a machine to move atoms

教机器移动原子

基本信息

批准号：
2434395
负责人：
金额：
--
依托单位：
University of Leeds
依托单位国家：
英国
项目类别：
Studentship
财政年份：
2020
资助国家：
英国
起止时间：
2020 至无数据
项目状态：
已结题

来源：
https://gtr.ukri.org/projects?ref=studentship-2434395
关键词：
Teaching machine move atoms

项目摘要

It is now routine in world leading laboratories to perform experiments where single atoms can be placed with atomic precision. These experiments are performed using scanning probe microscopes, and allow us to study the electrical and chemical properties of matter with sub-molecular, and even "sub-atomic" precision.However, these experiments rely on highly trained expert researchers operating machines by hand in an extremely time-consuming process. There is now significant interest in other fields towards developing machine learning algorithms, and autonomous systems, to allow machines to understand complex problems - for example facial recognition and self-driving vehicles.In this project the candidate will build on recent developments to develop automation and machine learning algorithms for integration with cutting edge scanning probe microscopes investigating intermolecular and interatomic interactions, essentially teaching a machine to "see" and move single atoms and molecules. This will be accomplished by direct interfacing with the Ultra-high resolution scanning probe microscopes currently housed in the MNP group at the University of Leeds.The candidate will develop expertise in both experimental physics and in programming (using Python and/or LabVIEW) and develop transferable skills working as part of a team of dedicated researchers in a challenging experimental environment.

现在，在世界领先的实验室里，进行实验，可以以原子精度放置单个原子，这已经是一种惯例。这些实验使用扫描探针显微镜进行，使我们能够以亚分子甚至“亚原子”的精度研究物质的电学和化学性质。然而，这些实验依赖于训练有素的专业研究人员手动操作机器，过程极其耗时。目前，其他领域对开发机器学习算法和自主系统有着浓厚的兴趣，以使机器能够理解复杂的问题-例如面部识别和自动驾驶车辆。在这个项目中，候选人将在最近的发展基础上开发自动化和机器学习算法，用于与尖端扫描探针显微镜集成，研究分子间和原子间的相互作用，本质上是教一台机器“看”和移动单个原子和分子。这将通过与利兹大学MNP组目前拥有的超高分辨率扫描探针显微镜直接接口来实现。候选人将发展实验物理和编程（使用Python和/或LabVIEW）方面的专业知识，并在具有挑战性的实验环境中作为专门研究人员团队的一员发展可转移的技能。