Robotic Repair (2024-07)¶

Through the deployment of a mobile construction robot capable of conducting high-resolution object scanning and precise in situ Additive Manufacturing (AM), we present a novel design-to-fabrication workflow for repairing existing building structures. The integration of AM techniques into context-aware mobile robotic systems enables high-precision in-place fabrication for new construction and for repair of existing structures. The benefits of transferring AM processes on-site extend in relation to tolerance handling, direct manipulation of existing structures, and removing constraints on shape stability compared to pre-fabricated elements by utilizing the context. By incorporating geometrical data obtained through 3D capture methods into the design and planning environment of architects and engineers, a direct interface between the existing building site and the planned digital geometry is created, facilitating accurate design of in-place repair or additions to existing building components. To evaluate this approach, we conducted an experiment in which a mobile robot equipped with a clay extrusion 3D printing system conceptually repaired a set of damaged brick wall segments. This workflow involved capturing the existing context with two levels of resolution: low-resolution 3D scene capture with a depth camera to generate a trajectory for high-resolution scanning, from which a dense point cloud is recorded using a 2D laser profile sensor by following the designated trajectories. This dense point cloud enables the operator to identify both the geometry of the existing brick wall, generate the missing volume, and a print path trajectory that fits the bounds of the volume while considering functional and architectural parameters. The accurate completion of the missing volume was successfully demonstrated by in-place 3D printing using clay extrusion with the mobile robotic system, showing the conceptual effectiveness of the proposed approach.