RAPID: Wireless Positioning for Mitigating COVID19 Surface Transmissions

点击翻译按钮获取中文摘要

The project develops a new micro-location technology for tracking hand movements and alerting users when they are about to touch their face. The technology uses wireless signals and mobile sensors to track the distance between a user's hand and their face, and introduces new algorithms and frameworks that enable achieving high accuracy and robustness at ultra-low cost. Such technology can significantly reduce COVID19 surface transmissions, which account for 10% of all COVID transmissions according to the CDC and scientific studies. The success of this project can have an immediate impact on essential workers (in factories and grocery stores) and aide in a quicker economic recovery while minimizing the impact of a potential second wave as the economy re-opens. The potential long-term impact of this research extends beyond COVID19 and future pandemics to providing transformative capabilities in networked micro-location for smart environments, indoor navigation, and asset tracking.The goal of this proposal is to design and build a low-cost (sub-$5), ubiquitous wireless positioning technology that allows tracking and predicting hand-to-face distance. Developing such a technology requires addressing challenges in terms of accuracy (centimeter-scale), interference (from co-existing technologies and multi-path reflections), and compatibility (with existing ubiquitous technologies). To overcome these challenges, the project introduces a principled multi-modal sensor fusion framework for mobile devices. This framework operates by fusing various sensing modalities that already exist in mobile devices -- including BLE, accelerometers, magnetometers, and ultrasound. Taken individually, ultra-low-cost sensors for each of these modalities lack the necessary accuracy and robustness for centimeter-scale positioning; however, because they experience uncorrelated sources of noise and interference, combining these modalities in a principled probabilistic framework enables achieving high accuracy and robustness while maintaining low cost. If successful, the resulting system would be the most accurate, low-cost, and ubiquitious micro-location technology to date.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.