Researchers at MIT’s Senseable City Laboratory report that crowd-sourced data from the smartphones of vehicle occupants crossing bridges can help monitor the structural integrity of bridges. An open access article about their demonstration has been published in the Nature journal communications engineering.
Monitoring and managing the structural health of bridges requires expensive specialized sensor networks. Over the past decade, researchers predicted that cheap ubiquitous mobile sensors would revolutionize infrastructure maintenance; Still, extracting useful information in the field with sufficient precision remains a challenge. Here we report the accurate determination of critical physical properties, modal frequencies, of two real bridges from everyday vehicle driving data.
– Matarazzo et al.
The team collected smartphone data from controlled field experiments and uncontrolled Uber rides on the Golden Gate Bridge — a long-span suspension bridge — and developed an analysis method to accurately recover modal properties.
Researchers also successfully applied the method to partially controlled crowdsourced data collected on a short-span highway bridge in Italy.
Illustrating the controlled data collection and spatial segmentation approach. a Sensor arrangement on the dashboard of the first vehicle (Nissan Sentra) used to record the first fifty trips. b Sensor arrangement on the dashboard of the second vehicle (Ford Focus) used to record fifty-two trips. All vehicle trips across the Golden Gate Bridge had their smartphones facing up, so one axis was well aligned with gravity. Such orientation is not essential; However, knowledge of the configuration of the sensors is helpful for data preprocessing. c General scheme of the spatial segmentation of a bridge defined by two independent parameters: Δs and c, which remain the same over the length of the bridge. The red circles represent the centers of each segment, while the light colored boxes show the segment widths. A close-up of three adjacent segments si-1sIand si+1is shown to detail the segmentation parameters: c is the length of each segment, cO is the length of overlap between segments and Δs is the distance between the centers (red circles) of adjacent segments. Matarazzo et al.
Smartphones, containing dozens of sensors, are worn by almost 50% of the world’s population. Recent applications of smartphones in construction have shown that smartphone accelerometers can adequately capture structural vibrations. Theoretical and experimental research on vehicle-bridge interaction relationships have established governing equations and influential parameters.
The researchers found that bridge vibration frequencies can be identified from smartphone-vehicle driving data under real-world conditions. Data from a single trip is insufficient; Still, only 100 crowdsourced datasets can provide useful modal frequency estimates (below 6% error) for both short-span and long-span bridges.
Overall, the analyzes of the controlled and ridesourced data (total N=174) yielded accurate estimates of ten (seven distinct) modal frequencies of the Golden Gate Bridge; five of them had an error of 0.000%. The number of trips considered in the primary study was less than 0.1% of the daily trips on the Golden Gate Bridge; This shows that smartphones around the world offer enormous detection potential, containing valuable information about bridges and other important infrastructure. In addition, the accuracy of the most likely modal frequencies (MPMFs) improved as the number of datasets increased.
– Matarazzo et al.
Further analysis predicted that incorporating crowdsourced data into a maintenance plan for a new bridge could extend service life by more than 14 years (a 30% increase) at no additional cost.
Our results suggest that massive and inexpensive datasets collected by smartphones could play a role in monitoring the health of existing transport infrastructure.
– Matarazzo et al.
Matarazzo TJ, Kondor D, Milardo S. et al. (2022) “Dynamic Monitoring of Crowdsourced Bridges Using Smartphone Vehicle Rides.” Municipal Eng 1, 29 doi: 10.1038/s44172-022-00025-4