IRoP Combines Augmented Reality and Additive Manufacturing for 3D Designer Plaster Murals
Building on a range of disciplines, from additive manufacturing to augmented reality, the Interactive Robot Plastering project impresses.
Researchers from ETH Zurich have put robots and humans in harmony for an unusual task: three-dimensional plaster murals, designed using augmented reality and applied by robotic plaster-sprayers.
The culmination of a PhD research project started in 2018, the team's work details what it has termed "Interactive Robotic Plastering," or IRoP — putting robotic arms fitted with plaster-spraying nozzles in concert with designers and suitably-skilled workers to quickly apply organic-looking plaster designs to walls to form three-dimensional murals.
"The research combines three elements," the team writes in the paper's abstract, brought to our attention by IEEE Spectrum. "Interactive design tools, an augmented reality interface, and a robotic spraying system. Plastering is a complex process relying on tacit knowledge and craftsmanship, making it difficult to simulate and automate. However, our system utilizes a controller-based interaction system to enable diverse users to interactively create articulated plasterwork in-situ."
The heart of the system is Robotic Plaster Spraying, an adaptive thin-layer printing technique roughly equivalent to an additive manufacturing 3D printer: Layers of plaster are built up into a three-dimensional volume without support structures or post-processing. To this, the team added an augmented reality interface based on projectors — using room-sized projection mapping to provide an interactive design system, which both visualizes the finished mural and allows for its modification pre-construction.
The hardware for this is, by and large, off-the-shelf: The interface is based on commercial projectors, a pair of laptops — one acting as a server for the Robot Operating System (ROS) and the other tracking HTC Vive wands used by the designers — while the construction is handled by a UR10 robotic arm fitted with spray gun, controller box, pump, and an external axis to allow it to travel along a wall.
"For skilled workers," the team concludes, "the user-study validates that IRoP enables those with limited robotic programming and design experience to manipulate complex design and robotic fabrication processes intuitively, particularly due to the gestural input. Thus, IRoP has demonstrated a high potential to lower the barriers for skilled workers to robotic fabrication technology and computational design."
"The project shows that by implementing intelligent and collaborative human-machine workflows, we can support a socially sustainable integration of robotic construction processes and expand our repertoire of potential material systems."
The team's work has been published under open-access terms in the Proceedings of the CHI Conference on Human Factors in Computing Systems 2022 (CHI '22).