Reconstructing a complete and accurate 3D representation of indoor construction scenes is an important step towards automated visual monitoring of construction projects. For fast access to construction’s as-built visual data, construction drones are programmed to autonomously navigate the outdoor space and collect the data. However, due to limited satellite signal indoors, ground rovers provide safer and more reliable autonomous navigation in the narrow cluttered indoor space. In this paper we present a novel pipeline for 4D BIM-driven mapping of the as-built state of indoor construction site. 2D Light Detection and Ranging (LiDAR) sensors are mounted on an Unmanned Ground Vehicle (UGV) for Simultaneous Localization and Mapping (SLAM). The developed method consists of (1) BIM-driven data collection planning; (2) automatic mission navigation; (3) LiDAR data collection and (4) dynamic obstacle avoidance. Experiments show the applicability of the developed data collection strategy and the improved safety of automatic mission execution using UGVs.