
Until recently, in-ground loop detectors were the option for detecting traffic at an intersection. These industry-standard devices are great for sensing large steel objects—like traditional motor vehicles—and providing years of maintenance-free service when installed correctly. But today, choosing a traffic detection method isn’t so simple.
Modern vehicles—especially electric vehicles—are built with less steel, making in-ground sensors less effective. And loop detectors have always had trouble recognizing bicyclists and motorcyclists. Replacing or adding new loop detectors to existing roads also requires cutting into the pavement, adding to the expense and reducing the lifespan of both the device and the road.
As technology advances, new video- and radar-based detection systems are becoming more readily available. As a result, transportation agencies now have alternatives to traditional loop detectors, but these nonintrusive systems are not always the right solution.
“Nonintrusive systems can have some benefits over traditional vehicle detection systems, but they also have drawbacks,” says Derek Lehrke, signal operations and construction engineer with the Minnesota Department of Transportation (MnDOT)’s Metro District.
In a MnDOT-sponsored project, U of M researchers identified factors for local agencies to consider when choosing between in-ground sensors and nonintrusive technologies (NITs)—allowing them to make better-informed decisions about which traffic detection system is right for them.
Researchers began their work by investigating NIT systems currently on the market and reviewing literature about those systems. Next, they interviewed traffic operations staff at Minnesota transportation agencies using NIT systems.
“The interviews provided valuable insight into the real-world experiences of Minnesota transportation agencies, including the benefits, challenges, and costs of operating and maintaining the different systems,” says Yao-Yi Chiang, an associate professor with the Department of Computer Science and Engineering (CEGE) and CTS Research Scholar. Chiang took over as the project’s principal investigator from John Hourdos, former CEGE research associate professor, now with the Federal Highway Administration.
The most-used NITs in Minnesota were evaluated in the Twin Cities at six intersections under a variety of conditions. Researchers discovered that weather and environmental factors can greatly affect NIT performance.

“Glare, snow, ice, and dirt can accumulate on the camera lens and limit a systems’ visibility and effectiveness,” Chiang says.
After comparing NIT products in the field, researchers found that no system clearly outperformed others in all weather conditions. However, some had fewer issues in certain scenarios.
The cost of NITs was also investigated, revealing that installation costs were lower for NITs than for traditional embedded loop detectors. For a sample intersection, researchers estimated that the cost of installing NIT devices was 1.5 times less than that of traditional loop detectors.
The project findings were used to develop a new decision tool to help operators select, install, and maintain the right NIT product for their needs. In addition to considering weather conditions when selecting an NIT system, other recommendations for improving NIT durability and performance include angling cameras below the horizon line, installing shields or filters to reduce glare, repositioning cameras when necessary, using a central monitoring system to perform checks and diagnose problems remotely, and installing heat shields on all NIT devices.
“Though NIT products and systems are continually evolving, our research highlights the core features so that city and county engineers across Minnesota can choose the system that suits their local environment and community,” Chiang says.
—Megan Tsai, contributing writer