RINGS: Resilient and Low-Latency Networks for Situation Awareness in the Factory of the Future

及时和准确的态势感知（例如，了解人、物体和车辆等代理的状态）是未来工厂自动化的关键因素。工业4.0范式预见了无数的异构代理，这些代理健壮可靠地沟通和协调它们的行动。FoF环境的传感和通信条件恶劣，导致信息延迟和数据不确定性。这可能会损害当前最先进的定位和决策程序，并要求对FoF网络中过时和不准确信息具有鲁强性的技术。该项目旨在通过为FoF环境设计延迟感知和延迟弹性技术，建立一个新的框架，彻底改变定位和决策的方法。该项目包括两个主要方面：延迟感知网络定位和多智能体FoF网络中的延迟弹性决策。首先，研究小组正在开发一个框架，用于设计延迟感知定位技术。这些技术通过调整准确性和延迟权衡以及融合从异构设备获得的数据，对破坏性事件具有鲁棒性。其次，研究团队正在开发具有延迟弹性的决策算法。这些算法通过有效地分配网络资源和协调代理加速学习来适应过时的信息。该项目将作为大规模多代理FoF网络的基础，该网络能够在存在延迟的高动态环境中运行。在这个项目中开发的方法对过时和不准确的信息是稳健的，为未来十年的全自动工业铺平了道路。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Timely and accurate situation awareness (e.g., knowing the state of agents such as people, objects, and vehicles) is a key enabler for automating the factory of the future (FoF). The Industry 4.0 paradigm foresees a myriad of heterogeneous agents that robustly and reliably communicate and coordinate their actions. FoF environments manifest harsh conditions for sensing and communication, resulting in information latency and data uncertainty. This may impair current state-of-the-art localization and decision-making procedures and calls for techniques that are robust to outdated and inaccurate information in FoF networks. The project aims to establish a new framework that revolutionizes the approach to localization and decision making by designing latency-aware and latency-resilient techniques for FoF environments. This project includes two main aspects: latency-aware network localization and latency-resilient decision making in multi-agent FoF networks. First, the team of researchers is developing a framework for the design of latency-aware localization techniques. These techniques are robust to disruptive events by adapting accuracy-latency tradeoffs and by fusing data obtained from heterogeneous devices. Second, the team of researchers is developing latency-resilient decision-making algorithms. These algorithms are resilient to outdated information by efficiently allocating network resources and by coordinating the agents for accelerated learning. This project will serve as a foundation for large-scale multi-agent FoF networks that are capable of operating in highly dynamic environments in the presence of latency. The methodologies developed in this project are robust to outdated and inaccurate information, paving the way to fully automated industries in the next decade.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.