Buried Pipes: Point Clouds × As-Built Management — Three Required Accuracy Levels and Checkpoints

In infrastructure pipeline construction for buried water and sewer pipes, gas pipes, and the like, "as-built management" to confirm that their positions and shapes match the design after completion is indispensable. Especially for piping buried underground, once it is installed it can no longer be seen directly, so it is necessary to measure and record it accurately before backfilling. Traditionally, it has been common to measure key depths and positions manually using tape measures, staffs (leveling rods), and levels, and to record them on paper drawings and photographic ledgers. However, manual surveying is labor- and time-intensive, and because it can only measure a limited number of points, it lacks comprehensiveness and carries the risk of overlooking construction errors.

One promising means to solve these issues is as-built management utilizing three-dimensional point cloud data. By using a collection of numerous points (point cloud) acquired by laser scanners or photogrammetry, the terrain and structures surrounding buried pipelines can be digitally recorded in their entirety. By thoroughly scanning the site, it is possible to capture shapes including areas that could not be measured by conventional methods, aiming for zero overlooked construction defects. In addition, because point cloud data can be freely measured and analyzed on a computer, comparison and verification against design values is efficient. In recent years, easy-to-use measurement technologies such as smartphone-embedded LiDAR and drone aerial photography have emerged, greatly lowering the barriers to point cloud use on construction sites.

This article explains the required accuracy when using point clouds for as-built management of buried pipes, dividing it into three typical tolerance ranges: ±30 mm (±1.18 in), ±50 mm (±1.97 in), and ±100 mm (±3.94 in). It focuses on the cases that require each level of accuracy and the key points to check on site. Furthermore, it covers the latest 3D measurement technologies—smartphone LiDAR, drones, terrestrial laser scanners, and RTK-GNSS—and introduces methods to efficiently acquire high-accuracy point cloud data and the key points for ensuring quality.

What Is As-Built Management for Buried Pipes?

First, we will clarify the fundamentals of as-built management for buried pipeline works. As-built management is the process of checking and recording whether the construction has been completed with the shapes and dimensions specified in the design. This process is important across civil and construction projects in general, but it is especially required for underground structures such as water and sewer pipelines and gas pipelines—because it is difficult to inspect them after burial, thorough as-built management during the construction stage is necessary. In public works projects, the results of as-built management often form conditions for inspection acceptance or for handover, and it is regarded as essential for ensuring quality.

The main items to check in as-built management of buried pipes include the pipe’s longitudinal slope (gradient), the pipe invert (pipe bottom) elevation, the pipe’s horizontal position (deviation from the design line), the clearance distance to adjacent buried utilities, and the cover (backfill thickness). For example, with sewer pipes it is important to ensure an appropriate slope, and strict measurement is required because even small elevation errors can lead to poor drainage. For water and gas pipes, it is necessary to confirm whether they are laid according to the planned position and buried at the specified depth in order to avoid interference with other pipes or structures. For power and communication cable ducts as well, maintaining accurate location information can reduce the risk of accidental damage during future excavation work.

Conventionally, these checks have been carried out by construction personnel measuring pipe diameters and clearances with tape measures and using a level to measure heights. The measurement results are then organized into as-built management drawings and ledgers with photos and submitted to the client. However, there were several issues with manual as-built measurements. Measuring large areas with limited personnel takes time, and because measurement points are limited to only a few locations, it is difficult to detect all anomalies. Also, during the busy work of backfilling buried pipes, ensuring that photos and records are taken without omission is burdensome, and there have been cases where forgotten photos or recording errors were discovered later and caused problems. As-built management is supposed to be the last line of defense to guarantee construction quality, but the conventional method has been plagued by the risk of human error and oversights.

Against this background, new technologies have been explored to enable reliable and efficient as-built management even in buried-pipe construction. Among them, the utilization of point cloud data described below has been raising expectations among on-site personnel.

Benefits of As-built Management Using Point Cloud Data

Point cloud data is a collection of numerous points acquired by laser scanning or photogrammetry that represents the surface shape of an object in three dimensions. By utilizing this point cloud, various benefits arise for as-built management of buried pipes.

First, it should be noted that point cloud measurement can record the site geometry exhaustively. Because it can preserve site conditions—such as the piping itself and the surrounding excavation surfaces—in both planar and three-dimensional forms, it enables capture of subtle sags and localized unevenness that would have been missed by manual measurements with only a few data points. By thoroughly scanning the entire pipe and its surroundings, the risk of overlooking construction errors in unmeasured areas can be greatly reduced.

Next, the introduction of 3D scanning leads to greater efficiency in measurement work and faster as-built inspections. Using laser scanners or drone photogrammetry allows large areas to be measured in a short time, greatly reducing the labor required for as-built measurements that traditionally mobilized several people and took half a day. For example, if a terrain model is generated from aerial photographs taken by a drone, even a pipeline extending several hundred meters can be fully recorded in a matter of tens of minutes. The acquired point cloud data can be compared with design data using dedicated software, enabling immediate on-site pass/fail determination of the as-built condition. By checking inspection results on the spot and immediately correcting nonconforming areas, later rework can be avoided and the construction schedule shortened.

Furthermore, another advantage is that digital records make reliable quality certification possible. Point cloud data and the photos linked to them serve as objective evidence of the as-built condition. Because measurement date/time and coordinate information are recorded automatically, there is no worry that someone will later doubt, "Was that part really measured?" The acquired 3D data can be shared with stakeholders via the cloud, making it easy for clients and inspectors to remotely check the as-built status. Compared with paper drawings and photo logs, information sharing becomes markedly faster and more reliable. Moreover, by utilizing digital data, tasks such as the automatic creation of as-built management documents and remeasuring arbitrary dimensions on the point cloud become easy, reducing the effort required to prepare reports.

In this way, as-built management using point clouds is expected to be a method that can achieve both "zero oversights" quality assurance and operational efficiency. The Ministry of Land, Infrastructure, Transport and Tourism is also promoting 3D as-built management that utilizes ICT through initiatives such as "i-Construction", and its application to buried pipe construction will likely continue to expand.

Required Accuracy for As-Built Management (±30 mm (±1.18 in), ±50 mm (±1.97 in), ±100 mm (±3.94 in))

Scanning a site with point clouds yields a vast amount of information, but for as-built management what matters is the accuracy of that data. No matter how detailed the point cloud, if measurement errors are large it cannot withstand comparison and verification against the design. In general, in as-built management of buried pipes, the required accuracy (tolerance) varies depending on the use and conditions. As a representative guideline, in strict cases an accuracy of about ±30 mm (±1.18 in) is required, in standard cases about ±50 mm (±1.97 in), and in relatively permissive cases about ±100 mm (±3.94 in). Below, we explain the expected situations and examples of suitable measurement methods for each of these three accuracy levels.

Cases requiring high precision of ±30 mm (±1.18 in) demand the strictest accuracy control. When securing pipeline gradient for sewer lines is critical, or in urban areas where existing structures are dense and pipe routes have clearance of only a few cm (a few in) from other equipment, particularly high accuracy is required from the measurement and construction stages. In sewer systems, an error of a few cm (a few in) affects drainage capacity, so the pipe invert elevation must be kept within ±30 mm (±1.18 in), and there are situations where centimeter-level positioning accuracy (cm level accuracy (half-inch accuracy)) is indispensable to ensure clearance from other buried utilities. In such cases, it is important to guarantee point cloud measurement accuracy within a few cm (a few in). Specifically, it is necessary to choose measurement methods whose instruments themselves have high accuracy, such as using terrestrial laser scanners or scanning with smartphone LiDAR augmented by high-precision RTK-GNSS for positioning enhancement. Furthermore, it is required to thoroughly remove and correct data errors by calibrating point clouds with known control points and cross-checking using multiple positioning methods.

Cases requiring an accuracy of about ±50 mm (±1.97 in; 5 cm (2.0 in)) are common, such as as-built management of water and gas pipes. Although small deviations do not immediately affect function, if the position differs greatly from the drawings it may hinder locating the pipe during future excavation investigations. Therefore, in practice it is desirable to record the buried positions with several-centimeter-level accuracy (cm level accuracy (half-inch accuracy)). If about ±50 mm (±1.97 in; 5 cm (2.0 in)) is acceptable, many point-cloud measurement technologies that have become common in recent years can readily achieve this. For example, terrain models from drone photogrammetry, when properly georeferenced, can achieve several-centimeter accuracy in both horizontal and vertical directions (cm level accuracy (half-inch accuracy)). Also, smartphone LiDAR scans, when combined with positional correction from high-precision GNSS, can limit errors to within 5 cm (2.0 in). However, if the measurement area is wide or satellite reception conditions are poor, errors tend to increase, so post-measurement verification—such as comparing with reference points to check for discrepancies—is necessary.

There are cases where a permissible tolerance of about ±100 mm (±3.94 in) is sufficient. Depending on the type of buried pipe and the purpose of the work, there are situations where a rough accuracy on the order of ±10 cm (±3.9 in) (a decimeter) is sufficient. For example, in cases such as drainage pipe work within private property or installation records of communication cable conduits, where public accuracy standards are not strict, it may be practically acceptable as long as the approximate position and elevation are known.

Also, in surveys that measure the approximate positions of aging existing pipes using point clouds, precision down to a few cm (a few in) is not required, and the objective can sometimes be achieved if the positional relationships can be grasped within an error range of about 10 cm (3.9 in). A precision of about ±100 mm (±3.94 in) is a range that can be relatively easily achieved even with relatively simple surveys using smartphones or drones. However, a 10 cm (3.9 in) deviation is by no means small.

In the future, when excavation relies on the positions of buried pipes for other works, if the records are off it can lead to accidents or rework. Although there is some leeway, it is desirable to leave data as close to the measured values as possible. Even for point cloud data obtained by simple methods, it is important to take measures such as later comparing with known points to check that there are no deviations of 50-100 mm (1.97-3.94 in) or more, and, if necessary, apply corrections.

Checkpoints for Ensuring Accuracy

When performing as-built management using point clouds, it is important to measure and process the acquired data appropriately so that it meets the required accuracy. Here, we introduce the main checkpoints to confirm on site to ensure point cloud accuracy.

Use of survey control points: First, it is fundamental to establish survey control points (control points) with known coordinates on-site and use them as the reference for point cloud data. No matter how high-performance the equipment is, if the overall coordinate reference is shifted, the as-built drawings are meaningless. Establish sufficiently accurate control points near the site in advance (existing public control points or project control points), and measure those points before and after point cloud surveying to assign correct position coordinates to the obtained point cloud. Compare the known distances between control points with the measured distances on the point cloud to check for errors, and if there are obvious shifts perform corrections; it is important to always take the attitude of verifying the spatial accuracy of the measurement data.

Appropriate equipment selection and calibration: It is also important to choose measurement equipment and methods that match the required accuracy. As mentioned above, in situations where ±30 mm (±1.18 in) is required, using high-precision equipment is indispensable, whereas for ±100 mm (±3.94 in) a simpler method may suffice. Depending on site conditions (outdoors/indoors, whether measurement is possible from the air, the presence of obstacles, etc.), use and switch between tools such as smartphones, drones, and ground-based laser scanners. Also, always perform equipment calibration. For example, for drone aerial imaging perform camera calibration in advance, and for smartphone LiDAR stabilize the sensor before starting a scan (e.g., reset the IMU) to eliminate factors that lead to accuracy degradation. It is also effective to compare and verify multiple measurement results to check for any systematic errors originating from the equipment.

Managing measurement range and point cloud density: In point cloud measurement, it is important to sufficiently scan the object from all directions so that no blind spots or unmeasured areas remain. For buried pipes, special attention must be paid to measurement omissions in parts that are hard to see from the surface, such as inside excavation trenches or on the far side of the pipe. If necessary, consider scanning from the opposite side or temporarily widening the excavation for the sake of measurement. Point cloud density (spacing between points) also affects accuracy. For dimensions that require high precision, measure at an appropriate resolution so that a sufficient number of points are obtained on the point cloud. For example, if you need to check pipe diameters or the shape around joints, scanning closely at high density to reduce the spacing between points makes it easier to later distinguish differences down to a few centimeters (a few in).

Noise removal and data processing: The acquired point cloud data may include noise points (erroneous measurements) caused by the characteristics of the equipment or the surrounding environment. It is necessary to appropriately filter these and remove outliers. Also, when integrating multiple measurement results, pay attention to the accuracy of coordinate alignment (registration). When overlaying point clouds obtained from different positions and angles, use corresponding feature points or place control points as targets to prevent misalignment. After integrating the point clouds, re-check the errors against known points and confirm they are within the required accuracy before proceeding to as-built drawing generation and numerical comparisons.

Comparison and verification against design data: Once point-cloud data has been acquired, always compare it with design drawings or 3D design data to verify the as-built condition. Predefine the points to check for deviations from design values, such as pipe alignment (planar position deviations), slopes and upstream/downstream end elevations, and clearance distances to adjacent structures. By generating cross-sections in point-cloud processing software or measuring thickness on arbitrary sections, you can quantify fine as-built differences that were previously difficult to detect. In recent years, methods that generate as-built inspection heat maps (color-coded visualization of elevation differences) from point-cloud data and allow instant identification of defective areas have been gaining popularity. The visual verification results provided by color-coding also help build consensus with the client. When nonconformities are found, promptly correct them on site and reacquire point-cloud data to confirm the post-correction condition in a cycle that ensures reliable quality assurance.

By carrying out point cloud measurement and data processing with the above points in mind, you can obtain high-quality as-built data that meets the required accuracy. Conversely, if you neglect these checks, you risk recording inaccurate data without noticing the errors. Because point cloud data can be measured easily, it is all the more important to rigorously adhere to basic accuracy management and quality checks.

Main technologies and equipment used for 3D measurement

There are various types of 3D measurement technologies that can be used for as-built management of buried pipes. Because each has different strengths and accuracy characteristics, it is important to choose between them based on site conditions and required accuracy. Here, we explain the characteristics and practical use points of four representative methods: smartphone LiDAR, drone aerial imaging, terrestrial laser scanning, and RTK-GNSS positioning.

Smartphone LiDAR measurement: Some recent smartphones and tablets are equipped with small LiDAR (Light Detection and Ranging) sensors. Using these, you can easily obtain surrounding 3D point cloud data with a dedicated app. The advantages of smartphone LiDAR are mobility and ease of use. You can scan while freely walking around a site with a device that fits in your pocket, and it is less likely to miss fine details inside narrow trenches or around piping. The measurement range is limited to a few meters (a few m / a few ft), but because you can approach the target and acquire high-density point clouds, it is also suitable for recording detailed shapes such as fittings and valves. The standalone accuracy of smartphone LiDAR measurements is said to be on the order of several centimeters to several tens of centimeters (several cm to several dozen cm / several in to several dozen in), but by combining corrections from high-precision GNSS and pre/post-known-point calibration, it can be improved to a level sufficient for as-built management. Above all, because it can quickly produce 3D records of complex shapes that are difficult to measure manually, it is an effective method for preserving detailed shapes before backfilling buried pipes.

Drone aerial photography (photogrammetry): Photogrammetry using small unmanned aerial vehicles (drones) has also become widespread as an efficient method for acquiring as-built data over wide areas. A camera mounted on the drone photographs the entire site from above, and software reconstructs point clouds and terrain models from the set of photos. The advantages of drone aerial photography are that it can cover a wide area at once and can record areas that cannot be accessed on the ground from a bird’s-eye view. Even long pipelines and large-scale land development sites can be turned into 3D models of the whole area in a short time. However, in photogrammetry, areas shaded by trees or structures do not appear in the data, so caution is needed. For buried pipes, the bottoms of deep excavation trenches and the sides of the pipes are difficult to capture from aerial photos, and supplementary measurements from the ground may be required. In terms of accuracy, if the drone is equipped with RTK‑GNSS or multiple ground control targets are placed on the ground and their coordinates measured, it is possible to keep errors in the range of several cm to about 5 cm (several in to about 2.0 in). It is particularly effective for applications that require large-area ground elevation verification and for as-built management of roadbeds and ground surfaces after burial, where the goal is to capture surface-level vertical differences across an area.

Terrestrial laser scanner: A ground-mounted laser scanner (TLS: Terrestrial Laser Scanner) set on a tripod or similar is a high-performance surveying instrument capable of acquiring high-density point clouds with millimeter- to several-millimeter (mm (in)) accuracy. It is a laser measuring device that sweeps up, down, left, and right, acquiring 360-degree point clouds of the surroundings in a short time.

The strengths of terrestrial laser scanners are their high accuracy and detailed shape reproduction. They can perform millimeter-level distance measurements (mm (in)) even for targets several tens of meters (m (ft)) away, making them ideal when you want to record an entire pipeline with high precision. Also, because they can acquire an enormous number of points — millions at a time — they can digitize even the subtle undulations of terrain and structures without omission.

On the other hand, because the equipment is large and expensive, operation requires specialized knowledge, and there are handling constraints in confined sites. Because many areas may be shadowed and not visible from a single location, measurements from multiple positions and the merging (registration) of point clouds are necessary. Therefore, terrestrial laser scanners are often used only for special cases that demand very high accuracy or for as-built management of large-scale, complex shapes that are difficult to measure by other methods.

In typical pipeline construction, it is more practical to handle most tasks with easier methods such as smartphone or drone surveys, and to perform additional high-precision scans only at key locations.

Use of RTK-GNSS positioning: The RTK method of GNSS (satellite positioning) is a technology that achieves high-precision position measurement with errors of a few centimeters (a few in). By using dedicated GNSS receivers and receiving correction information from a base station in real time, it can reduce the common GPS positioning error of a few meters (a few ft) down to a few centimeters (a few in). In buried pipe construction, RTK-GNSS is often used alone for as-built measurements. For example, one method is to measure coordinates with RTK at each pipe start/end and bend point and record the difference from the design values. Recently, cases combining the aforementioned point-cloud measurement and RTK-GNSS have increased. Equipping drones or smartphones with RTK-GNSS allows high-precision absolute coordinates to be assigned directly to the acquired point clouds, simplifying post-processing alignment and reducing the effort for accuracy verification. Caution is still required in densely built-up urban areas or mountainous regions where satellite reception is difficult, but in Japan augmentation signals from the Quasi-Zenith Satellite System have been implemented, making it easier to obtain high precision even with standalone positioning in open-sky environments. RTK-GNSS can now be considered an indispensable enabling technology as the "foundation of positioning" for point clouds.

Easily achieve high-precision as-built management with LRTK

Finally, I will touch on a new as-built management method that has attracted attention in recent years: smartphones + high-precision GNSS. Thanks to advances in technologies such as the smartphone LiDAR and RTK-GNSS mentioned above, centimeter-class (cm) positioning and 3D measurement, which previously required specialized equipment, are becoming achievable on smartphones that fit in anyone’s hand. A representative example is the system called "LRTK," which consists of a pocket-size RTK-GNSS receiver that can be attached to an iPhone and a dedicated app. By simply attaching a small antenna to the smartphone, positioning error can be reduced to a few centimeters (a few in), and those high-precision coordinates can be immediately assigned to point clouds acquired by the smartphone’s built-in LiDAR scanner or camera.

By using a system like LRTK, it becomes possible to scan the as-built condition of buried pipes on site while walking, and check deviations from the design data on the spot, without spending time installing control points or performing point-cloud registration in post-processing. In fact, advanced inspections—such as AR overlaying the design model and the acquired point cloud on a smartphone screen to check even minute height differences—can now be handled with a single smartphone. Acquired point cloud data and measured coordinates can also be shared instantly via the cloud, enabling automatic output to inspection documents and reuse for later maintenance (for example, even after paving, displaying the buried pipe locations as a transparent overlay in smartphone AR to prevent accidental excavation). By digitally replacing as-built management that previously relied on manpower and experience and bringing a system to the field that lets anyone measure accurately, the simple surveying system using smartphone + GNSS can be considered a groundbreaking solution.

High-precision, efficient as-built management directly contributes to both quality assurance and productivity improvement. By leveraging advanced tools like LRTK, even limited personnel can reliably record and inspect the as-built condition of buried pipes, creating valuable digital assets for future infrastructure maintenance. The benefits that the integration of point cloud measurement technology and satellite positioning technology brings to job sites are substantial, and it is expected that such practices will increasingly become standard across construction and surveying sites. Consider actively adopting the latest technologies to elevate the accuracy and efficiency of buried pipe as-built management to the next level.