UAV-Enabled Site-to-BIM Automation: Aerial Robotic- and Computer Vision-Based Development of As-Built/As-Is BIMs and Quality Control

摘要

The automated integration of as-built and as-is conditions into building information models (BIM) remains a primary challenge for unoccupied aerial vehicles (UAV)-enabled facility and construction inspection. Due to the lack of site-to-BIM data pipelines supporting reality capture technologies such as camera-mounted UAVs, BIMs lose their effectiveness over time since they do not accurately reflect the as-built and as-is conditions; the UAV-captured visual data also remains underutilized. This paper proposes and demonstrates a novel industry foundation classes (IFC)-based solution for UAV-enabled as-built and as-is BIM development, quality control, and smart inspections. It first identifies the requirements of UAV-enabled site-to-BIM automation and its framework; then, it elaborates on the proposed solution which aims to: (1) enable automated conformance checking and quality control using data queried from BIMs and UAV-captured reality in the form of images; and more importantly; and (2) enable the integration of on-site captured reality, including the as-built and as-is conditions, into BIMs. The method leverages the non-proprietary IFC schema, empowering OpenBIM applications, and facilitating interoperability, a core challenge in the information modeling domain. In addition to its support for UAV-enabled applications, the technique can provide the same site-to-BIM functionalities for other types of reality capture and other robotic data collection efforts during construction, commissioning, and facility management.