Artificial intelligence is being harnessed by voice-controlled personal assistants, chatbot financial services, and even smart thermostats—now the University of Waterloo is applying algorithms to improve an age-old profession: bricklaying.
Researchers at the university used AI software to study how masons position their body during bricklaying, revealing new insights into the safest poses and most productive way to work through machine learning.
“The people in skilled trades learn or acquire a kind of physical wisdom that they can’t even articulate,” said Carl Haas in a statement. Hass is a professor of civil and environmental engineering at Waterloo. “It’s pretty amazing and pretty important.”
The study was published in Automation in Construction today and analyzed 21 masons with varying levels of expertise. The researchers found that master masons don’t follow standard ergonomic rules taught to novices, identifying the techniques those with more expertise use to limit the loads on their joints.
While Haas found experts use fewer and more ergonomically safe poses, they are in fact more productive and use less energy without wasted motions.
“They’re basically doing the work twice as fast with half the effort–and they’re doing it with higher quality,” said Haas. “It’s really intriguing.”
Tracking movements with sensor suits and using a machine learning algorithm to classify masons’ poses, the researchers determined that master masons put less stress on their bodies and had the tendency to swing rather than lift blocks.
“Skilled masons work in ways we can show are safer, but we don’t quite understand yet how they manage to do that,” said Hass. “Now we need to understand the dynamics.”
Pulling findings from two studies, the researchers determined patterns of body positions that can now be applied to develop affordable mason training systems. The researchers are planning to do a more in-depth study of how mason experts move on the job, and are also developing a training system that uses sensor suits to give trainees immediate feedback.