Drone and 3D Measurement Changing On-Site Preliminary Quantity Design

Introduction: The Importance of Preliminary Quantity Design and On-Site Challenges

*Figure: On-site surveying with a drone. The appeal lies in being able to survey high or wide areas safely and efficiently.* In construction projects, preliminary quantity design is an important process in the planning stage for estimating the amount of earthwork and the scale of land development required for the work. Accurate preliminary quantity calculations directly affect cost estimates and schedule planning and can determine the success or failure of a project. However, at the early design stage, detailed survey data are often lacking on site, and quantities must be estimated from limited information. Therefore, conventional preliminary quantity design has always been accompanied by anxiety over whether the estimates will “significantly deviate from actual quantities,” posing headaches for designers and clients alike.

On-site conditions and constraints are also challenges. In terrain where surveying is difficult, such as steep slopes in mountainous areas or wetlands, it is inherently hard to obtain a sufficient number of survey points by manual methods, and the accuracy of preliminary quantities tends to suffer. Also, when covering a wide area with limited personnel, point intervals inevitably become coarse, risking the overlooking of local depressions or terrain changes. Because of these challenges, it was not uncommon in the past to assume conservatively large values or apply heuristic corrections during preliminary quantity design. As a result, unnecessary budget appropriations or contract changes due to shortages occurred, indicating room for improvement in traditional methods.

Limitations of Conventional Methods and the Dilemma of Accuracy vs. Effort

Traditional preliminary quantity design relied mainly on manual topographic surveys and measurements on drawings. For example, a designer might extract cross sections from paper topographic maps or existing 2D CAD data and calculate earthwork volumes. However, estimating volumes from contour lines on paper maps has limits, and it was practically difficult to ensure accuracy due to coarse elevation points and reading errors.

Even when conducting field surveys, many problems remained. Covering a large site with total stations or GPS surveying requires significant manpower and time. For example, for a site on the scale of several hectares, a survey team might spend multiple days acquiring numerous survey points, followed by in-house data organization and volume calculations. The number of points that can be obtained manually is limited, and if one tries to finish quickly, one must increase the interval between survey points, leading to only a rough understanding of the terrain. Consequently, there were risks such as “overlooking earth volume in unmeasured valleys” or “underestimating quantities due to missed unevenness.”

The more you try to improve accuracy, the more effort increases. Detailed surveying requires more days and personnel, raising costs, so some degree of compromise was necessary at the preliminary design stage. In this way, the limitations of conventional methods include the difficulty of balancing accuracy and efficiency. Busy sites could not afford sufficient time, and preliminary estimates were often produced with “accuracy as a secondary concern.”

How 3D Models Are Generated by Drone Surveying

A technology that has recently been transforming this situation is photogrammetry using drones. By capturing numerous high-resolution photos from the air with a drone-mounted camera and analyzing them with dedicated software, it is possible to generate a detailed 3D model of the site. This process uses a method called SfM (Structure from Motion), which extracts feature points from overlapping areas of multiple photos and analyzes their positional relationships to create a point cloud. Further MVS (Multi-View Stereo) processing densifies the point cloud, enabling smooth terrain representation. The result is 3D point cloud data, which represents the surface as a collection of countless points.

What used to be measured point by point by hand can now be obtained as tens of millions of survey points in a short time with drone photogrammetry. For example, drone flight time is on the order of tens of minutes, and post-flight data processing is increasingly automated, allowing a precise terrain model to be completed in a matter of hours. Recent high-performance drones are equipped with RTK-GNSS and can be expected to provide position information with centimeter-level accuracy (half-inch accuracy). Therefore, by only supplementarily setting a minimal number of control points (described later), one can generate point cloud models with high accuracy aligned to the site coordinate system. Aerial 3D modeling dramatically shortens the days previously required for surveying while providing detailed data that was previously unavailable.

Mechanism for Instant Calculation of Earthwork, Cross Sections, and Areas from Point Cloud Data

*Figure: Example of terrain point cloud data obtained from drone photogrammetry. The ground surface is densely represented by countless points.* Point cloud data obtained by drone photogrammetry are essentially a digital copy of the actual site. Each point has X, Y, Z coordinate values (and color information based on the photos), so various measurements can be made by analyzing the point cloud. If you generate a mesh or contour lines from the ground surface point cloud on dedicated software, earthwork calculations and cross-section creation that were previously done manually become dramatically more efficient.

For example, in calculating earthwork volumes, you overlay the ground surface model generated from the point cloud with the planned development surface and compute the volume difference between the two. This instantly yields how much fill or cut is required for a given development area. Because mesh comparisons are performed directly from the point cloud, calculations are highly accurate even for complex or irregular terrain, significantly reducing errors compared to traditional manual calculations.

It is also easy to generate cross sections along arbitrary lines. By specifying a “survey line” on the point cloud data, you can immediately draw longitudinal and transverse profiles along that line. Where previously one would connect field-measured points to create cross sections, now it can be done at the push of a button. Orthophotos (top-down images corrected for distortion) are also generated, allowing area measurements on plan views and overlaying with design drawings to check as-built conditions. In short, with point cloud data you can digitally measure all geometric information such as distances, areas, and volumes.

These processes are semi-automatically handled by dedicated software or cloud services, so the person in charge only needs to review the results and make adjustments as necessary. For example, fill-and-cut volume calculation results are immediately displayed in tables and graphs, and multiple cross sections can be extracted consecutively and output in bulk. This is also effective for as-built inspections: by comparing the point cloud data of the completed terrain with the planned design surface, you can visualize in a color-coded heatmap whether the fill has reached the specified elevation and where there are surpluses or shortages. This changes as-built verification from on-site eyeballing or using a scale to an objective validation based on data.

Case Study: Preliminary Design Work Reduced from 3 Days to Half a Day Using Drones

As an example demonstrating the power of drone surveying and point cloud analysis, consider preliminary quantity design work at a planned development site. The site is about 10 hectares, and conventionally a surveying team of 2–3 people would collect data by ground surveying, and the design staff would calculate volumes. In this case, field reconnaissance and surveying would take 2–3 days, and an additional day would be needed for in-house data organization and calculation, totaling 3–4 days to compile the preliminary quantities.

Switching to drone surveying dramatically shortened the work time. Preparation and flight for aerial photography could be completed in half a day, and the analysis from photos to point cloud model generation and volume calculation was finished the same day. As a result, work that used to take 3 days was effectively completed in half a day. Time was reduced to about 1/6, and manpower was cut to less than half since surveying could be done by one drone operator and one assistant.

This speed enables quick feedback to clients. With early access to accurate quantity information, decisions on design changes and budget measures can be considered earlier. Also, short turnaround means repeated surveying becomes easy. If needed, re-shooting the current conditions by drone allows comparison of multiple development scenarios before construction starts. Where previously surveying could be done only once due to time and cost constraints, drones make it easy to perform surveys whenever conditions change.

Ability to Handle Sites That Are Difficult to Survey (Slopes and Inaccessible Areas)

Drone surveying is particularly beneficial for dangerous sites that are difficult for people to access. On steep slopes or unstable slopes where collapse is a risk, traditional ground surveying required personnel with safety lines and carried a high risk of missed measurements and human error. With drones, you can obtain detailed data of such hazardous locations from a safe distance by flying the aircraft overhead. Even targets that were difficult to measure directly, such as talus slopes or quarry faces, can be captured and their shapes understood by converting aerial photos into point clouds.

Drones are also invaluable immediately after disasters or for inspecting aging infrastructure. For example, right after a large landslide, access may be impossible due to aftershocks or secondary disaster risks, but drones can scan the disaster area from above. This enables early preliminary quantity estimation (such as the volume of collapsed soil or urgent fill requirements), contributing to faster initial response.

Moreover, an increasing number of inspections and measurements of high structures like bridges and dams are being done by drone without erecting scaffolding or using elevated work platforms. For locations that are difficult to capture with a drone camera alone, such as underwater areas or under tree canopies, combining ground-based laser scanners or UAV laser surveying (LiDAR-equipped drones) is becoming common. In this way, drones and 3D measurement technologies make it possible to measure places that were previously unmeasurable, dramatically improving on-site safety and data accuracy.

Further Improvement of On-Site Coordinate Accuracy by Combining with LRTK

Point cloud models generated by drone photogrammetry have very high relative shape accuracy, but they may not automatically guarantee precise absolute coordinates. To align point clouds to site-specific coordinate systems (such as plane rectangular coordinate systems or public coordinates), it is necessary to place several reference points (GCP: Ground Control Points) on site and provide their high-precision positions as known points. LRTK is highly effective for surveying these reference points.

LRTK (a smartphone RTK solution provided by Reflextia) uses an ultra-compact RTK-GNSS receiver that attaches to a smartphone, enabling easy centimeter-class positioning *Figure: The pocket-sized RTK-GNSS receiver “LRTK Phone” attached to a smartphone. This device allows anyone to perform high-precision positioning easily.*. Traditionally, RTK positioning required expensive dedicated equipment and skilled operators, but with LRTK the era of one person, one smartphone for high-precision surveying has become a reality. By combining a smartphone with a small device and launching a dedicated app, you can obtain coordinates of any point in real time simply by pressing a button.

If you provide control points measured with LRTK to the drone-acquired point cloud, you can align the entire point cloud to a public coordinate system with high accuracy. This allows the resulting 3D data to be overlaid precisely with design drawings and other survey data. In addition to single-point surveying, LRTK can be used for point cloud scanning in conjunction with a smartphone’s camera and AR functions. For example, simply walking around the site with a smartphone can acquire a point cloud with absolute coordinates, which is useful for surveying indoor spaces or the underside of bridges where drones cannot fly. Combining drones and LRTK enables high-accuracy coverage of all points across a site, further improving the data accuracy required for preliminary quantity design.

Conclusion: 3D Measurement Changes Preliminary Accuracy and Work Styles

As described above, drone-based 3D measurement is attracting attention as a technology that dramatically improves the accuracy and efficiency of preliminary quantity design on site. Earthwork calculations that once took days can now be performed almost in real time, and dangerous surveying tasks are being replaced by safer methods. As a result, accurate quantity information can be obtained from the design stage, enabling rational planning based on data rather than conservative safety margins or heuristic judgments. This offers significant benefits to both clients and contractors, optimizing budgets, reducing the risk of contract changes, and even reducing environmental impact (by cutting unnecessary excavation and transport).

Changes are also occurring in work styles. With reduced surveying burden, engineers can obtain results in shorter time and devote freed-up time to design review and other tasks. There is less need to carry heavy surveying equipment through hazardous sites; instead, the trend is shifting toward analyzing and utilizing point cloud data at desks. For young engineers, working with digital tools is attractive, which positively impacts recruitment and training. Indeed, the DX (digital transformation) brought by 3D measurement is simultaneously realizing improved accuracy in preliminary quantity design and reforms in how people work.

Going forward, as the Ministry of Land, Infrastructure, Transport and Tourism advances i-Construction initiatives, the use of such 3D surveying technologies is expected to spread further. If these methods become the standard on sites, it will be possible to manage projects digitally from the preliminary stage through completion, greatly improving productivity and transparency in construction projects.

Appendix: Smartphone × High-Accuracy Survey Workflow Starting with LRTK

Finally, we touch on the smartphone × high-accuracy surveying workflow, which is attracting attention alongside drone surveying. By utilizing the aforementioned LRTK, your handheld smartphone instantly becomes a high-accuracy surveying device. The basic on-site usage is simple.

• Attach device and start up: Attach the LRTK Phone receiver to your smartphone and launch the dedicated app. After initial setup to receive augmentation signals (such as Japan’s QZSS “Michibiki” CLAS signal), you are ready.

• Point measurement: At the point you want to measure, simply tap the button on the smartphone screen to record the coordinates (latitude, longitude, and elevation) of that location. Measurement is instantaneous, and accuracy depends on conditions but is generally within a few centimeters (a few in) for both horizontal and vertical. You can also save the measured points with names or notes.

• Cloud sharing: Recorded data can be uploaded to the cloud on site. There is no need to connect cables or manually organize data back at the office. On the cloud you can view measured points on a map and share data among multiple people for real-time use.

In this way, simple surveying using LRTK is overwhelmingly easier and faster than conventional surveying equipment. For example, field leveling surveys that used to require multiple people can in some cases be replaced by a single LRTK-equipped smartphone. Its high mobility makes it a “daily-use surveying tool” that lets you measure whenever needed, transforming on-site work styles. Combining wide-area 3D measurement by drone with pinpoint high-precision positioning by a smartphone is like adding an iron club to a demon—an unbeatable combination. If you are about to undertake preliminary quantity design, consider actively incorporating these latest tools. A new workflow that balances efficiency and improved accuracy will strongly support your projects.