Automating surface flatness control using terrestrial laser scanning and building information models

Frédéric Bosché, Emeline Guenet

    Research output: Contribution to journalArticlepeer-review

    125 Citations (Scopus)
    679 Downloads (Pure)

    Abstract

    Current practice in the control surface flatness requires a significant amount of time and labour, and delivers results based on few sample measurements. Developments of Terrestrial Laser Scanning (TLS) and Building Information Modelling (BIM) offer great opportunities to achieve a leap forward in the efficiency and completeness of dimensional control operations. This paper presents an approach that demonstrates the value of this integration for surface
    flatness control. The approach employs the Scan-vs-BIM principle of Bosche and Haas (2008) [1] to segment TLS point clouds acquired on-site, by matching each point to the corresponding object in the BIM model. The novel approach
    then automatically applies two different standard flatness control techniques, Straightedge and F-Numbers, to the TLS points associated to each floor, and concludes with regard to their compliance with given tolerances. The approach is tested and validated using data from two actual concrete slabs. Results conrm the suitability of using TLS for conducting standard dimensional controls, and validate the performance of our system when compared to traditional measurements methods (in terms of both quality and efficiency). Furthermore, a novel straightedge generation method is proposed and demonstrated that enables more complete and homogeneous analysis of floor flatness for insignificant additional processing times.
    Original languageEnglish
    Pages (from-to)212-226
    Number of pages15
    JournalAutomation in Construction
    Volume44
    DOIs
    Publication statusPublished - Aug 2014

    Fingerprint

    Dive into the research topics of 'Automating surface flatness control using terrestrial laser scanning and building information models'. Together they form a unique fingerprint.

    Cite this