Particularly in the execution and operation phase, little research has been done on how data and information can be automatically acquired and integrated into building information models (BIMs). The presented approach thus includes autonomous data acquisition with UAVs (unmanned aerial vehicle), semantic information extraction and an update of existing BIMs. All steps shall be automated to enable real-time information delivery in short cycles. Based on the 4D-Monitoring-BIM (4D-MBIM), relevant components are identified and waypoints for the drone-based data acquisition are derived. Subsequently, the UAV mission is planned based on the 4D- MBIM, considering the data acquisition technology, possible obstacles and other boundary conditions. The UAV mission and quality of acquired data is evaluated within a full simulation of the data acquisition mission. The data acquisition is carried out by means of autonomously operating UAVs. Within the acquired point clouds, a semantic segmentation using Machine Learning methods is per- formed to detect objects. Per object class, single objects are clustered and subsequently (i) a full BIM reconstruction, (ii) BIM to scan comparison and (iii) further information extraction e.g., concrete crack and spalling detection, is performed. All information will be represented according to the BIM methodology using open standards.