Driver smartphones could help predict (and prevent) bridge collapse
Monitoring the structural integrity of heavily used and aging bridges is complex work that often involves arrays of costly sensors, but new research has demonstrated how smartphones might one day do the heavy lifting. Scientists have shown how an app can be used to detect subtle vibrations that indicate the likelihood of bridge collapse, planting the seeds for a low-cost, crowdsourced solution to maintaining such all-important infrastructure.
In order to track the health of a bridge, engineers will tune into natural vibrations known as modal frequencies, which can shift over time to reveal changes in the structure’s integrity. This can be achieved through sensors such as accelerometers that are placed on the bridge, though we have seen how wireless sensor systems could offer a cheaper way forward, and also how autonomous drones could streamline these types of operations.
The authors of the new study have instead sought to tap into the existing sensor systems in everyday smartphones. The team built an Android-based app specifically to gather smartphone accelerometer data on modal frequencies from vehicles passing over bridges. The idea was to put them to work and see how well the data matched up with that collected by a traditional set of bridge-monitoring sensors.
This was tested out on the Golden Gate Bridge, with the researchers driving over the bridge 102 times with the app running, and also asking Uber drivers to do the same over 72 trips. The data was then compared to modal frequency data from a set of 240 traditional sensors that had been attached to the Golden Gate Bridge for three months. The team found that the phone data closely mirrored the data gathered by the traditional sensors, offering a close match for 10 types of low-frequency vibrations engineers look at in these instances. In five cases, there was no difference at all.
“We were able to show that many of these frequencies correspond very accurately to the prominent modal frequencies of the bridge,” said study author Paolo Santi, principal research scientist at MIT’s Senseable City Lab.
The Golden Gate Bridge is a suspension bridge, a type that accounts for only 1% of all bridges in the US. To expand on these findings and explore how the technology might apply to smaller concrete span bridges that make up around 41 percent, the team turned to a bridge of this type in Ciampino, Italy.
This part of the study involved 280 vehicle trips over the bridge with the smartphone app onboard, with the data compared to that from a set of six sensors that had been attached to the bridge for seven months. Though the researchers found a divergence in the data on certain modal frequencies of up to 2.3%, they are enthusiastic about the potential, as this was an improvement on the 5.5% divergence observed in a smaller data sat. This suggests expanding the data further through more trips could further hone the accuracy of the technique.
“We still have work to do, but we believe that our approach could be scaled up easily – all the way to the level of an entire country,” said co-author Carlo Ratti. “It might not reach the accuracy that one can get using fixed sensors installed on a bridge, but it could become a very interesting early-warning system. Small anomalies could then suggest when to carry out further analyses.”
The team also looked at how incorporating this kind of smartphone data into maintenance plans for bridges could impact their longevity, and calculate that mobile-device monitoring could add between 15 and 30% extra years to their service life.
“These results suggest that massive and inexpensive datasets collected by smartphones could play an important role in monitoring the health of existing transportation infrastructure,” the authors write.
The research was published in the journal Communications Engineering.