When bridges are inspected for cracks and other defects that could lead to their collapse, engineers must either hang beneath those bridges on lines, or view them from elevated platforms. Whichever approach is used, a lot of setup is involved, and defects may get missed. In the future, however, unmanned aircraft may be able to more quickly and thoroughly check out bridges, working with wireless sensors built into the structures.
The system is being developed at Massachusetts' Tufts University, by assistant professors Babak Moaveni and Usman Khan.
UPGRADE TO NEW ATLAS PLUS
More than 1,200 New Atlas Plus subscribers directly support our journalism, and get access to our premium ad-free site and email newsletter. Join them for just US$19 a year.UPGRADE
Five years ago, Moaveni installed a series of 10 wired sensors on a 145-foot (44-m)-long footbridge on the Tufts campus. These sensors measured vibrations that passed through the bridge, caused by people walking across it.
He subsequently placed about 5,000 pounds (2,268 kg) of concrete weights on the bridge, to simulate the sort of stress that could lead to its collapse. This extra weight caused a change in the nature of the pedestrian-caused vibrations, which was easily detected by the sensors.
The scientists are now working on a way of scaling that system up, so that it could be applied to larger, car-carrying bridges. In that scenario, a series of wireless sensors located on the bridge's beams and joints would continuously process and record vibrational data. Autonomous quadcopter drones would regularly fly out and hover beside each of those sensors, wirelessly receiving data from them while also snapping photos of that part of the structure.
The aircraft would send that sensor data and imagery to a central computer, which would use an algorithm to detect any changes in vibrations that could be a result of a defect in the concrete. When such changes were detected, engineers would be alerted, and could check the photos of the corresponding parts of the bridge.
According to Moaveni, the system should already be capable of detecting severe damage, but still needs some tweaking before it can pick up on more subtle defects.
Source: Tufts University