Sami Masri, a professor of Civil and Environmental Engineering at the University of Southern California, Los Angeles, uses the Microsoft Kinect to gather data on the condition of roads. The device’s infrared projector, infrared camera and RGB camera are mounted beneath a vehicle to create a 3D image of the road as the vehicle drives; this data along with GPS information is used to identify cracks and potholes and localize them.
If devices like Masri’s prototype were installed in all 2016 Toyota Corollas, for example, live road health monitoring could be possible for Departments of Transportation (DOT’s) across the nation, similar to live traffic monitoring in apps like Google Maps and Waze—a community-updated traffic app.
What bars this application from becoming a reality isn’t the technology.
“The price of these sensors has dropped by an order of magnitude in a few years,” says Masri, referring to the now sub $100 Kinect. He says the technology to do powerful structural health monitoring is here. Processing the Big Data from thousands of cars about millions of cracks and potholes across miles of roadway, is the most technologically difficult part of bringing the idea to life. In Masri’s study his team created a defect detection and quantification system based on many previous studies and local and federal pavement management guidelines.
Masri’s study pits his automated pavement defect detection process against the current manual and semi-automated processes being used today. Manual data collection is expensive and can be dangerous to inspectors and can lead to defects due to the subjectivity by the rater. In semi-automated systems—which are also expensive—the process of data acquisition is carried out using vehicles capable of data collection equipped with cameras or laser sensors, says his study. But the pavement condition part of the assessment is still a manual process that is prone to inspector opinion.
Masri’s system automates the collection and the processing of data, eliminating subjectivity of an inspector and streamlining the process.
Two other roadblocks impede mass implementation. Before a crowd-sourced DOT road health app appears in the Mac App Store, industry partnership, and investment are necessary, Masri says.
Investment might be warranted.
“Driving on roads in need of repair costs U.S. motorists $67 billion a year in extra vehicle repairs and operating costs,” states a study by The Road Information Program (TRIP), a national transportation research group. The TRIP study goes on to state that road conditions play a major role in the 200,000 annual traffic fatalities as well.
If perfected and implemented, Masri’s system has the potential to help DOT’s across the nation better use the $70.3 billion per year is allocated for highways’ condition improvement. With research like Masri’s available, maybe it isn’t only more funding that is needed for proper infrastructure management, but investment in the right technology that could make real time, community-based, structural health monitoring a reality.