This thesis improves the planning and execution of formwork assembly for walls and columns—a repetitive, time-critical, and labor-intensive activity that strongly influences concrete quality, project cost, and schedule. The work implements a BIM-to-ROS pipeline that starts from a Revit-authored formwork model, exports panel instances via IFC, and transforms them into (i) a Neo4j property graph capturing panel identifiers, target poses, and precedence relationships, and (ii) simulation-ready assets by generating per-panel meshes and SDF models for Gazebo. These data products are then consumed by a ROS 2 execution layer that selects the next eligible panel from the graph, dispatches an installation goal to an InstallPanel action server, and writes back status updates for consistent progress tracking. By reducing the effort needed to prepare and adjust execution plans, improving consistency, and making progress easier to track, the approach supports more reliable delivery and helps the industry respond to skilled labor shortages, productivity losses, schedule delays, and safety risks associated with heavy, repetitive handling. The approach is evaluated in Gazebo Fortress using a mobile manipulation setup (Husky base, 6-DoF arm, vacuum gripper) in representative wall and column scenarios. The results show that BIM-derived targets and graph-based sequencing can drive automated goal dispatch with closed-loop feedback, achieving millimeter- and sub-millimeter placement accuracy for successfully installed panels within a 5mm translational tolerance band. Looking ahead, such data-driven workflows are a step toward construction sites where robots increasingly take on structured, repeatable tasks—such as moving, positioning, and assembling standardized components—while humans focus on supervision, quality control, and handling exceptions, enabling safer and more predictable production in the built environment.     «

This thesis improves the planning and execution of formwork assembly for walls and columns—a repetitive, time-critical, and labor-intensive activity that strongly influences concrete quality, project cost, and schedule. The work implements a BIM-to-ROS pipeline that starts from a Revit-authored formwork model, exports panel instances via IFC, and transforms them into (i) a Neo4j property graph capturing panel identifiers, target poses, and precedence relationships, and (ii) simulation-ready asse...     »