Automated integration of as-is point cloud information with as-planned BIM for interior construction

Abstract

Integrating 3D point cloud data in BIM can enhance the understanding and visualization of the actual state of the interior construction. We propose an overall process of registering as-is point cloud data to a BIM model and integrating as-is information to generate a 4D as-is BIM. The point cloud data of a portion of a complex multi-building construction is automatically registered in the planned BIM using an automated room floor plan search and matching algorithm. The as-is information obtained from the semantic features of the point cloud is integrated with the planned BIM. The BIM resulting from the integration is enriched with information about the construction status of elements at the specific time of the data collection. The proposed method was demonstrated on a real construction site, where point cloud data was collected four times, capturing the construction progress over about a month. The search and matching algorithm accurately matched all the sample rooms on the planned BIM model with the room shapes obtained from respective as-is point clouds. Furthermore, the as-is information was effectively extracted from the raw point cloud data instances and automatically integrated into the as-planned BIM.