Purdue steps up in construction robotics, automation

Researchers at Purdue University's Polytechnic Institute are working with automation software and robotics to try to improve and advance construction processes, according to an article published on the university’s website.
Jiansong Zhang, assistant professor of construction management technology, is working with algorithms that can support future automation on the jobsite in an effort to help continue the advancement in robotics technology that can help from the early stages of planning to the processes of laying rebar for concrete and more.
Zhang's research is focused on extracting information through algorithms that help computers review a building’s design model and compare it with city's and state's building codes, for example. Any code violations in the design are flagged for correction. "Even an average laptop, rather than a supercomputer," he said, "can perform the reasoning thousands of times faster than a human."

More firms and universities are researching ways in which robotics and automation can help spur productivity growth and advances in the industry, which is notorious for dragging in those regards compared to other industries. For instance, researchers at the Swiss Federal Institute of Technology in Zurich earlier this year developed a new timber construction method using robotics that they said allows complex wood structures to be built more easily.

Meanwhile, Construction Robotic's bricklaying semi-automated mason (SAM) robot is making headlines for its impressive work on jobsites. Using a metal robotic arm to spread mortar and then a laser-guided system to place the bricks, the robot is laying the bulk of the bricks at the $35.5 million University Arts Building at the University of Nevada, Reno. SAM also made its mark on a Hoar Construction federal project in Roanoke, Virginia, where it laid 250,000 — or 70% — of the project’s bricks.

Construction robotics are becoming smarter, too. Autodesk’s BrickBot is programed with 3D models, and can erect structures out of Legos to show how robots can perform complex tasks. The robot uses industrial robotic arms outfitted with cameras and sensors to generate data that fuel the machine's algorithm-based neural networks, allowing it to process and analyze results of what it is doing to adapt to its environment. Researchers hope to scale the technology for factory and construction applications.

Purdue certainly isn’t the only higher education institution researching robotics. Yale University researchers developed what they call “Robotic Skins” technology, which they claim can essentially turn everyday objects into robots. Although not designed with a specific task in mind, the researchers assert the skins could be used as wearable technology in complex tasks.