1. Home
  2. RIP
  3. View Record
https://trb-annual-meeting.nationalacademies.org/registration

Utilizing daily traffic as a sensor network for infrastructure health monitoring

Mobile sensing is a novel paradigm that offers numerous advantages over conventional stationary sensor networks for real time bridge monitoring. Mobile sensors have low setup costs, collect spatio-temporal information efficiently, and require no dedicated sensors to any particular structure. Most importantly, they can capture comprehensive spatial information using few sensors. The advantages of mobile sensing combined with the ubiquity of smartphones with internet of things (IoT) connectivity have motivated researchers to consider smartphones carried within vehicles as large-scale sensor networks that can contribute to the health assessment of structures. A practical implementation of mobile sensors has several challenges. Most notably, the signals collected within a vehicle's cabin is contaminated by the vehicle suspension dynamics and the road profile; therefore, the efficient extraction of bridge vibration from signals collected within the vehicle is of great importance. The majority of available approaches for addressing this are typically system specific and restricted by assumptions of linearity. This limits the scope of application since vehicles mostly act nonlinearly depending on their manufacturing specifications. In addition, the variety of vehicle systems and road conditions complicates the exploration for a unified method for this task. This project proposes deep learning frameworks with domain adaptability that enable vehicle signal decontamination in a more reliable and practical manner. This framework will transform vehicles into robust and high-quality vibration sensors for infrastructure monitoring. Furthermore, this will render smartphone-based vehicle sensing data a valuable source of information that will enable crowdsourcing and facilitate infrastructure condition assessment in real time at an unprecedented scale, rate and resolution.