RAPID: Low-cost, Batteryless Smart Personal Protective Equipment (PPE) Tackling the COVID-19 Pandemic

医护人员依靠口罩和其他个人防护装备(PPE)来保障他们的健康。新冠肺炎疫情表明，仅靠个人防护装备不能保护工人，因为疲劳、活动，甚至接触人都会影响工人的健康和能力。该项目的目标是开发智能个人防护设备，其中包括设计小型、低成本和可连接到面具上的智能无电池传感器设备。这些口罩将为工作人员提供有用的信息，以减少体积和成本，在不需要充电或维护的情况下自我管理健康，以不引人注目的方式保护工人。研究任务集中在建立一个能量收集和无电池硬件和软件平台，该平台可以在动态能量输入导致间歇性电源故障的情况下，对面罩内传感器数据进行连续推理和通知。本论文主要研究三个方面的工作：(1)开发间歇供电的人体感知硬件平台；(2)为内存受限的间歇系统开发一个基于任务的自适应检查点系统，以执行设备上的推理；(3)探索连续推理的自适应机制，以调节预测的估计精度以减少电源故障。最后，将研究产品集成到单一平台上，并为智能个人防护设备开发传感应用程序。本研究将为在新冠肺炎大流行和未来健康危机中使用智能个人防护设备为医护人员提供低成本、主动保护的计算手段。通过留下电池，这些设备可以以更小的尺寸和成本运行，无需维护，不需要充电。这些设备将向工作人员和医院提供可操作的数据和通知，以控制传染病的传播并保持一支随时准备好的、健康的劳动力队伍。这项研究为超受限计算系统的设计提供了参考，并可应用于保护其他基本工作人员和普通民众。开发的系统，包括软件、硬件、文档和详细的组装说明，将向社区提供。所有研究产品将在以下网站上提供：http://kamoamoa.com/projects/smart-ppe/.该网站将在工作完成后至少一年内保持可访问。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Healthcare workers rely on face masks and other personal protective equipment (PPE) to safeguard their health. The COVID-19 pandemic has shown that PPE alone cannot protect workers, because fatigue, activity, even exposure to people impact the health and ability of the worker. The goal of this project is to develop smart PPE, which includes the design of small, low cost, and smart batteryless sensor devices that can be attached to masks. The masks will provide useful information for workers to self-manage health without the need for recharging or maintenance at reduced size and cost, unobtrusively protecting the worker.The research tasks center on building an energy harvesting and battery-free hardware and software platform that conducts continuous inference and notification on in-mask sensor data despite intermittent power failures stemming from dynamic energy input. Three research tasks are pursued (1) prototyping a hardware platform for intermittently powered human sensing, (2) developing a task-based adaptive checkpointing system for memory-constrained intermittent systems that perform on-device inference, (3) exploring adaptive mechanisms for continuous inference which modulates estimated accuracy of prediction to reduce power failures. Finally, the research products are integrated into a single platform and sensing applications are developed for smart PPE.This research will enable the computational means for low cost, active protection of healthcare workers in the COVID-19 pandemic and future health crises using smart PPE. By leaving batteries behind, these devices can function maintenance-free, without recharging, at a reduced size and cost. These devices will provide actionable data and notifications to workers and hospitals for controlling the spread of infectious diseases and maintaining a ready, healthy workforce. The research informs ultra constrained computing system design and could be applied to protect other essential workers and the general population.The developed systems, including software, hardware, documentation, and detailed assembly instructions will be made available to the community. All research products will be made available at the following website http://kamoamoa.com/projects/smart-ppe/. The site will remain accessible for at least a year after the completion of the work.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.