Architects and engineers test robot’s capacity to build complex structures

Photo: Texas A&M College of Architecture

Boston — Working with an industrial-scale robot and polystyrene blocks at Autodesk’s BUILD Space in Boston, a team of architects and engineers from the Texas A&M College of Architecture are developing a full-scale, light-weight vault structure from interlocking and uniquely shaped structural modules.

This “geofoam” module technology could potentially be used for robotic construction of affordable housing, emergency shelters, field medical clinics and even support structures, said Amir Behzadan, associate professor of construction science and project lead of the Texas A&M team.

“The project was motivated by a technology-informed initiative to impart practical knowledge of robot-assisted construction to students and educators,” said Behzadan. “It demonstrates how design, construction, analysis, and planning teams can come together under one roof to deliver a project on time and within budget.”

As part of the six-week project that began last May, the group is programming a robot to create a unique, self-standing design that facilitates easy shipping and fast and simple assembly and disassembly, said team design lead Negar Kalantar, assistant professor of architecture.

“Because we are using a robot-mounted wire cutter, the design is limited to ruled surfaces,” she said. “The challenge is matching the pieces together in an arch so they each come out of one block of foam and there is no waste of material.”

Kalantar and Alireza Borhani, an architecture lecturer, guided the fundamental design strategy for the project. Both faculty members are interested in the applications of additive manufacturing technology, like 3-D printing, as a catalyst of innovation.

The team had to overcome structural analysis obstacles presented by the project’s modular nature, explained Julian Kang, an associate professor of construction science who conducted a finite element analysis of the vault. To conduct the evaluation, Kang said, “the output of the structural analysis of one module had to be used as the input of the analysis of the adjacent module, in an iterative process.”

After the project wraps in late June, the team will return to share its experience and acquired knowledge with students and faculty at Texas A&M, providing valuable insight into the rapidly developing world of architectural and construction robotics.