Light Detection and Ranging in Building Information Modelling

[edit] Introduction

Light Detection and Ranging (LiDAR) is a remote sensing technology that uses laser light to measure distances to physical surfaces. In building and construction, LiDAR can rapidly capture the geometry of existing buildings, structures, sites and building services, producing a dense set of spatial measurements known as a point cloud. The resulting data can be used to support surveying, three-dimensional computer-aided design (CAD), Building Information Modelling (BIM), refurbishment, conservation and construction verification.

LiDAR is an active sensing technology because it emits its own laser energy rather than relying on ambient light. Depending on the equipment and application, distances can be calculated from the time taken for laser pulses to return to the sensor or from changes in the phase of the emitted and returned signal. The technology can capture large quantities of spatial data without requiring direct physical contact with the surveyed surfaces.

[edit] LiDAR data acquisition

LiDAR systems can be deployed in different ways according to the scale, accuracy and accessibility requirements of a survey. Terrestrial laser scanning (TLS) uses equipment positioned at fixed locations, commonly on tripods, to capture the surrounding environment. Multiple scan positions can be used to obtain coverage of areas that would otherwise be obscured.

Mobile laser scanning systems carry the scanner through or around the area being surveyed. They may incorporate inertial measurement units and positioning systems to determine the position and orientation of the scanner as it moves. Mobile systems can be useful for surveying large or linear areas and for capturing extensive building interiors.

Airborne laser scanning can also be used to capture buildings, structures and surrounding terrain. Equipment mounted on aircraft or unmanned aerial vehicles can survey areas that are difficult to access from the ground, although the accuracy and suitability of such systems depend on the equipment, flight conditions, survey methodology and requirements of the project.

The positioning of scanners is important because LiDAR measurements generally require a direct line of sight between the sensor and the surface being measured. Multiple overlapping scan positions can therefore be required to minimise occluded areas and provide sufficient coverage.

[edit] Point clouds and registration

The principal output from a laser scanning survey is a point cloud, consisting of a large number of measured spatial points. Points commonly contain X, Y and Z coordinates and may also contain information such as laser intensity or colour derived from an associated camera. A point cloud represents measured surfaces but does not, by itself, constitute a BIM model or contain the semantic relationships associated with BIM objects.

Where multiple scan positions are used, the individual datasets must normally be registered into a common coordinate system. Registration can use overlapping areas of geometry, survey control points or targets established during the survey. The resulting dataset may also be georeferenced to an established coordinate system where this is required by the project.

Point cloud processing can include filtering noise and unwanted points, removing duplicate or erroneous measurements, classifying data and reducing dataset density where appropriate. Processing should preserve sufficient information to meet the accuracy and resolution requirements of the intended application. Excessive decimation can remove useful geometric information, while unnecessarily dense datasets can increase storage and processing requirements.

The accuracy of a point cloud depends on factors including the scanner, measurement range, surface characteristics, environmental conditions, scanner position, registration methodology and survey control. Accuracy should therefore be established from the specifications and quality assurance procedures applicable to the particular survey rather than assumed from the density of the point cloud alone.

[edit] LiDAR and BIM

LiDAR data can be used as a reference for creating or updating BIM models, particularly for existing buildings. In a scan-to-BIM workflow, the registered point cloud is used by a modeller to interpret existing geometry and construct appropriate BIM elements such as walls, floors, roofs, columns, beams, doors, windows, ducts and pipes.

The conversion from a point cloud to a BIM model is not simply an automatic transformation of points into intelligent building elements. It generally involves interpretation of the measured geometry and the application of agreed modelling rules. The resulting BIM elements can contain both geometric and non-graphical information, depending on the information requirements of the project.

The required level of information and geometric accuracy should be established before modelling begins. Different projects may require different levels of detail, and not every feature visible in a point cloud needs to be modelled. Modelling unnecessary detail can increase production time and model size without providing corresponding benefits.

LiDAR-derived information can be particularly useful where existing drawings are unavailable, incomplete or unreliable. Comparison of a point cloud with a design model can also assist in identifying differences between designed and constructed conditions.

[edit] Applications in construction and refurbishment

LiDAR is particularly useful in refurbishment, alteration and conservation projects because it provides a measured record of existing conditions. It can capture irregular geometries, variations in wall surfaces, structural deformations and building services that may not be accurately represented in existing drawings.

For heritage buildings, laser scanning can provide a detailed geometric record that can support condition assessment, conservation planning and documentation. Features such as complex façades, mouldings and other irregular architectural details can be recorded without requiring extensive manual measurement.

During construction, LiDAR can be used to verify installed work against design information, monitor progress and record as-built conditions. Point clouds can also be used to support coordination between architectural, structural and building services information, including the identification of potential spatial conflicts.

LiDAR can also support facilities management by providing an accurate spatial record of an existing asset. However, the usefulness of the resulting information model depends on the scope and accuracy of the survey, the quality of the modelling and the information requirements established for the asset.

[edit] Limitations

LiDAR is fundamentally a line-of-sight measurement technology. Solid walls and other opaque objects prevent the laser from measuring surfaces behind them, while areas concealed by ceilings, equipment, furniture or other obstructions may not be captured. Multiple scan positions can reduce, but not necessarily eliminate, these occlusions.

Some surfaces can also present measurement difficulties. Highly reflective, transparent or translucent materials can produce incomplete or unreliable measurements. Glass, polished surfaces and standing water can therefore require particular consideration during survey planning.

Large point clouds can require substantial storage capacity and processing resources. Point cloud viewing and manipulation can also affect software performance, particularly when several datasets are combined. Appropriate data management, segmentation and processing can help to control these issues.

LiDAR should not be regarded as a substitute for all other forms of surveying or investigation. A point cloud records the surfaces that have been measured, but it does not necessarily reveal concealed construction, material composition, structural condition or other information that cannot be determined from surface geometry. Additional surveys, inspections or investigations may therefore be required.

The transition from LiDAR data to BIM also introduces potential sources of error. The BIM model is an interpretation of the measured data, and discrepancies can arise through registration errors, incomplete survey coverage, modelling assumptions or inappropriate levels of detail. Quality assurance should therefore include comparison of the model against the source survey data where accuracy is important.

[edit] Related articles on Designing Buildings

• Laser scanning for building design and construction
• Point cloud
• Scan to BIM: Enhancing Building Information Modeling with 3D Laser Scanning
• BIM Modelling from Point Cloud Data
• Laser Scanning for 3D Modeling in Building Information Modeling
• Scan to BIM Workflow and Deliverables: A Technical Guide
• Scan to BIM for Historic Buildings
• BIM Dimensions: From 3D to 7D and Beyond
• Lasers in construction
• Building Information Modelling BIM