Keywords: building information modelling, laser scanner, photogrammetry, object recognition, geo-computational co-planar surfaces, spin images.

This paper explains the state-of-art techniques and approaches for modelling of existing structures such as a three-dimensional laser scanner. The research approach to building information modelling (BIM) for existing structures is introduced. A case study is elaborated to demonstrate how to produce three-dimensional CAD models of existing structures using a semi-automated technique.

Capturing and modelling three-dimensional information of the built environment is very challenging. Information models of real world data should be multifunctional, interoperable, intelligent, and multi-representational to be economical and sustainable. Interoperability means that different models and model parts may be smoothly integrated within spatial data infrastructures and may easily collaborate with systems from the domains of computer aided architectural design (CAAD) and computer graphics, which denotes intelligent semantic three-dimensional building models where objects know about their structure, thematic properties, and interrelationships with other objects. Special challenges arise from the need to manage multiple representations of entities in a three-dimensional city and building models concerning different scales, evolution over time, and concurrent versions. The new model quality also demands for advanced acquisition tools. These have to cope with the extraction of semantically meaningful features from images and laser scanning data, scale transitions, geometric and topological consistency, and international standards.

Building information modelling is the term used to describe a range of discipline-specific software applications that support all phases of the project lifecycle from conceptual design and construction documentation, to coordination and construction, and throughout ongoing facility management, maintenance, and operations. A BIM comprises individual building, site or GIS objects with attributes that define their detailed description and relationships that specify the nature of the context with other objects [1].

The case study building called Jactin House, under refurbishment in East Manchester, England, was scanned with a Riegl LMS Z210 scanner together with the companion software called RiSCANPro. For scans registration and post processing, RiSCANPro and the Polyworks software were used consecutively. Once all the scans are registered, a point cloud model of the scanned object is obtained. Following that, paper explains how this data is processed using co-planar surfaces to produce a three-dimensional CAD model and attaching semantic information to this CAD model to generate BIM model for existing structures.

This information modelling will enable automatic and fast data capture and enrichment for not only urban design and planning but also disaster management, environmental analysis, assessment and monitoring, GIS implementation, sophisticated simulation environments for different purposes such as climate change, regeneration simulation for complexity and uncertainty and so on.

1

Ballesty S., "Building Information Modelling for Facilities Management", project report by Co-operative Research Centre (CRC) for Construction Innovation, Queensland, Australia, URL, 2007.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £120 +P&P)