Sprayed Concrete Automation : A Unique and Complete Digital Fabrication Method
Worldwide, the sprayed concrete industry faces common challenges such as the need to optimize the process to reduce losses associated with material rebounding from the surface during placement, as well as a growing shortage of qualified operators due to the harsh job-site conditions. These challenges are at the heart of the research efforts of the Shotcrete Lab in Université Laval. Indeed, automation of sprayed concrete placement was seen as a perfect approach to address the economic and environmental impacts of rebound losses, and the challenging placement conditions found in structural shotcrete applications with large quantities of reinforcing bars.
Strong of a 6-axes industrial robotic arm acquired in 2019 that allows accuracy and repeatability, research progressed in two areas in parallel: the capturing of relevant digital information to facilitate the work of the operator and ultimately fully automate the process, and the methodical study of the effect of placement parameters (angles, distance, velocity, motion, etc.) on rebound losses.
After investigations on the effect of placement parameters, the exploration of the automation has come. The research made use of two cameras, one observing the orientation and the distance of the nozzle compared to the substrate, and a second observing the wall and the reinforcing bars for thickness control. The processed data is useful for both the shotcreter and the robot.
Probably the most significant from a material rheology point of view is the control of the placement parameters offered by the robotic arm, as opposed to human operator, allowing to devise a unique placement pattern that was found to reduce rebound losses by half. The implications of what was coined as the “fundamental rebound” opened a new research avenue investigating rheological properties during placement, offering a completely new insight in mixture designs optimization and spraying parameters optimization.
Preview
Cite
Access Statistic
