Researchers develop safety monitoring system for construction sites

The ViPER+ system can accurately track workers’ locations on job sites to enhance safety.

workers at construction site

sirisakboakaew | stock.adobe.com

Computer scientists at the University of Houston have developed a new monitoring system to keep construction workers safe at job sites.

To help contractors follow location-based safety policies—which often define safe areas for workers and equipment or define a safe distance when equipment is operating—the researchers created ViPER+, a system that automates the monitoring of these policies and detects any violations of the policies while workers and equipment are working. ViPER+ uses ultra-wideband technology for location tracking.

“The point of our research project was to enhance [the] safety of workers and equipment on a construction site by tracking their location,” says Alireza Ansaripour, a University of Houston computer science doctoral student and co-author of the study. “By tracking their location, we can monitor location-based policies related to the safety of workers and equipment in construction sites.”

The team says its ViPER+ system surmounts the challenges of other ultra-wideband-based real-time safety monitoring systems primarily because it overcomes non-line-of-sight situations. These are instances in which trucks, construction loaders and other equipment block the signal between the transmitter and receiver in ultra-wideband radio transmissions.

Ansaripour and his colleagues implemented a correction method in their localization, or location tracking algorithm to reduce the error caused by non-line of sight.

ViPER+ is an updated and improved version of the group’s initial system, ViPER. The greatest difference between the two is the enhanced location tracking on ViPER+, which is more accurate in non-line-of-sight situations.

The team tracked locations through tags and anchors. Tags are small ultra-wideband radio transmitters that are mounted on workers and vehicles to monitor their locations. Anchors are ultra-wideband receivers that receive signals from tags. The researchers then collected data from anchors to their computer server and estimated the location of vehicles and people in a construction site.

They tested their system twice in actual construction zones in Houston that were cordoned off for their experiment. But instead of real construction workers, students had to play the part.

“In our evaluation, all four construction workers had tags mounted. We also had one vehicle, either a truck or bulldozer with multiple tags on it and another static vehicle was used to create a non-line of sight situation,” Ansaripour says.

The first evaluation was in 2019 when researchers set up tags in an area of about 8,600 square feet, called the tracking zone. Four students operated as workers in the tracking zone while Ansaripour was managing the data flow of the system and made sure the experiment ran smoothly. In 2022, a similar scenario was set up but at a different construction site.

“Alireza is one of those students with brilliant ideas and the work ethic to see these ideas to fruition,” says Omprakash Gnawali, associate professor of computer science at the University of Houston’s College of Natural Sciences and Mathematics and co-author of the study. “Having that combination is important to get these technical projects to be successful.”

Future improvements

Future changes to the system include ironing out user design issues such as alerting construction workers when they are too close to moving machinery.

“We also have an issue creating a tracking zone that covers all of a construction site, not just a portion of it,” Ansaripour says. “There are still some improvements that need to be made for this to become a commercial product, but our work provides insight on how a real-time safety monitoring system can be used for safety tracking in construction sites.”

Other authors of the study include the University of Houston’s Milad Heydariaan, and from the University of Nebraska-Lincoln, corresponding authors Kyungki Kim and Hafiz Oyediran.

The research project was funded through the National Cooperative Highway Research Program (NCHRP) Idea of the National Academy of Sciences under the award NCHRP-206.