NextG无线网络与计算智能结合的工业数字孪生模型综述
Industrial Digital Twins at the Nexus of NextG Wireless Networks and Computational Intelligence: A Survey
摘要：
通过融合最新的通信和控制技术、计算和数据分析技术以及模块化制造，工业4.0通过网络物理系统（CPS）和数字孪生（DT）促进了网络物理世界的整合，以实现工业过程的监控、优化和自动化。DT是一个新兴的概念，但与CPS不同。与CPS一样，DT依赖于通信来创建对象或物理过程的高度一致、同步的数字镜像。此外，DT还在此精确图像上使用内置模型，通过反馈来模拟、分析、预测和优化其实时操作。随着工业物联网（IIoT）、边缘和云计算、机器学习、人工智能和高级数据分析领域的最新进展，DT正在各行业中迅速扩散。然而，现有的文献缺乏识别和讨论的作用和要求，这些技术在DT使能的行业从通信和计算的角度。在本文中，我们首先介绍了DT在智能行业中的功能、吸引力和创新用途。然后，我们通过系统地回顾和反思下一代（NextG）无线技术（例如，5G及更高网络）、各种工具（例如，信息时代、联邦学习、数据分析），以及联网计算中的其它有希望的趋势（例如，边缘和云计算）。此外，我们讨论了DT部署策略在不同的工业通信层，以满足工业应用的监测和控制的要求。我们还概述了几个关键的思考和未来的研究挑战和方向，以促进工业DT的采用。
英文摘要：
By amalgamating recent communication and control technologies, computing and data analytics techniques, and modular manufacturing, Industry~4.0 promotes integrating cyber-physical worlds through cyber-physical systems (CPS) and digital twin (DT) for monitoring, optimization, and prognostics of industrial processes. A DT is an emerging but conceptually different construct than CPS. Like CPS, DT relies on communication to create a highly-consistent, synchronized digital mirror image of the objects or physical processes. DT, in addition, uses built-in models on this precise image to simulate, analyze, predict, and optimize their real-time operation using feedback. DT is rapidly diffusing in the industries with recent advances in the industrial Internet of things (IIoT), edge and cloud computing, machine learning, artificial intelligence, and advanced data analytics. However, the existing literature lacks in identifying and discussing the role and requirements of these technologies in DT-enabled industries from the communication and computing perspective. In this article, we first present the functional aspects, appeal, and innovative use of DT in smart industries. Then, we elaborate on this perspective by systematically reviewing and reflecting on recent research in next-generation (NextG) wireless technologies (e.g., 5G and beyond networks), various tools (e.g., age of information, federated learning, data analytics), and other promising trends in networked computing (e.g., edge and cloud computing). Moreover, we discuss the DT deployment strategies at different industrial communication layers to meet the monitoring and control requirements of industrial applications. We also outline several key reflections and future research challenges and directions to facilitate industrial DT's adoption.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
