Drones and 3D Measurement Transform On‑Site Approximate Quantity Design

Introduction: The Importance of Approximate Quantity Design and On‑Site Challenges

*Figure: A site survey by drone. Its appeal lies in being able to survey high or wide areas safely and efficiently.* In construction projects, approximate quantity design is a critical process in the planning phase for estimating the amount of earthwork and the scale of site development required. Accurate calculation of approximate quantities directly affects cost estimates and schedule planning, and can determine a project’s success or failure. However, at the early design stage site survey data are often insufficient, forcing estimators to infer quantities from limited information. Consequently, traditional approximate quantity design has been plagued by persistent concerns—chiefly whether estimates will diverge significantly from actual quantities—causing headaches for designers and clients alike.

Site conditions and constraints are also problematic. In terrains that are difficult to survey, such as steep mountain slopes or wetlands, it can be inherently hard to obtain sufficient survey points by manual methods, which tends to sacrifice accuracy in quantity estimates. When limited personnel must cover large areas, point spacing inevitably becomes coarse, raising the risk of overlooking local depressions or terrain changes. Because of these issues, it was common in traditional approximate quantity design to assume conservatively large values or apply heuristic corrections. As a result, unnecessary budget allocations or contract changes due to underestimates occurred, indicating room for improvement in conventional methods.

Limitations of Traditional Methods: Accuracy vs. Labor Tradeoffs

Traditional approximate quantity design relied mainly on manual topographic surveys and measurements from drawings—for example, designers extracting cross sections from paper topographic maps or existing 2D CAD data and calculating earthwork volumes. But estimating volumes from contour lines on paper maps has limitations: coarse elevation control points and reading errors make it difficult to ensure accuracy.

Field surveying also presents numerous challenges. Covering a wide site with total stations or GPS surveying requires significant manpower and time. For a site of several hectares, a survey team might need days to collect many points, followed by in‑house data整理 and volume calculations. There is a limit to the number of points that can be collected manually; if trying to shorten time, point spacing must be increased, resulting in only a coarse understanding of the terrain. This can lead to risks such as "overlooked volumes in unmeasured valleys" or "undetected unevenness causing underestimated quantities."

The harder you try to improve accuracy, the more labor increases. Detailed surveys require more days and personnel, raising costs, so some compromise is unavoidable at the approximate design stage. In this way, a key limitation of traditional methods is the difficulty of balancing accuracy and efficiency. Busy sites often cannot spare the time, and estimates are frequently produced with "accuracy as a secondary concern."

How Drone Surveying Generates 3D Models

A technology that has recently begun to change this situation is photogrammetry using drones. By taking numerous high‑resolution aerial photos with a drone‑mounted camera and analyzing them with dedicated software, detailed 3D models of the site can be generated. This process uses a method called SfM (Structure from Motion), which extracts feature points from overlapping photos and analyzes their spatial relationships to create a point cloud. MVS (Multi‑View Stereo) processing then densifies the point cloud, enabling smooth terrain representations. The result is a 3D point cloud dataset that represents the ground surface as a collection of countless points.

What used to be measured point by point by hand can now be acquired as tens of millions of points in a short time with drone photogrammetry. For example, drone flight time may be only tens of minutes, and post‑processing has become increasingly automated, allowing a precise terrain model to be completed in hours. Modern high‑performance drones often include RTK‑GNSS, providing centimeter‑level positional accuracy during flight. Therefore, with only a minimal number of control points (discussed below) used as auxiliary references, it is possible to generate a high‑accuracy point cloud aligned with the site coordinate system. Aerial 3D modeling drastically shortens the days traditionally required for surveying while providing detailed data that were previously unavailable.

How Point Cloud Data Instantly Produce Volumes, Cross Sections, and Areas

*Figure: Example of terrain point cloud data obtained by drone photogrammetry. The ground surface is densely represented by countless points.* Drone photogrammetry point cloud data are essentially a digital replica of the site. Each point has X, Y, Z coordinates (and color information derived from photos), so various measurements can be made by analyzing the point cloud. Generating meshes or contour lines from the ground point cloud in dedicated software dramatically streamlines the earthwork calculations and cross‑section creation that used to be done manually.

For earthwork volume calculation, for instance, a surface model generated from the point cloud can be overlaid with the planned design surface, and the volume difference between the two is computed. This makes it possible to instantly determine how much to fill or cut in a given development area. Because the comparison is performed on meshes derived directly from the point cloud, accuracy is high even for complex or irregular terrain, significantly reducing errors compared to manual calculations.

Cutting cross sections along arbitrary lines is also easy. By specifying a "survey line" on the point cloud, longitudinal and transverse profiles along that line can be drawn immediately. What used to require connecting field‑measured points to produce cross sections can now be completed at the push of a button. Orthoimages (aerial photos corrected for distortion to create a true overhead view) are also generated, enabling area measurements on plan views and overlaying with design drawings to verify as‑built conditions. In short, with point cloud data, geometric information such as distance, area, and volume can all be measured digitally.

Because these processes are semi‑automated in dedicated software or cloud services, the operator only needs to review the results and make adjustments as necessary. For example, fill and cut volume results are displayed immediately in tables and graphs, and multiple cross sections can be extracted and output in batch. This is also effective for as‑built inspection: by comparing the completed terrain’s point cloud to the planned design, you can visualize whether fill has reached the specified elevation and where surpluses or deficits exist using color‑coded heat maps. This turns what used to be visual, tape‑based on‑site checks into objective, data‑driven verification.

Case Study: Reducing an Approximate Design Task from Three Days to Half a Day Using Drones

To illustrate the power of drone surveying and point cloud analysis, consider a case of approximate quantity design for a planned development of roughly 10 hectares. Traditionally, a 2–3 person survey team would acquire data by ground surveying, and a designer would calculate earthwork volumes. In this scenario, field reconnaissance and surveying might take 2–3 days, with another day for in‑office data整理 and calculations, totaling 3–4 days to compile approximate quantities.

Switching to drone surveying dramatically shortened the work. Drone preparations and flight could be completed in half a day, and the processing from photos to point cloud generation to volume calculation finished the same day. As a result, a three‑day task became effectively half a day. Time was reduced to about 1/6, and manpower was halved or better—surveying could be done by one drone operator and one assistant.

This speed enables rapid feedback to clients. Early access to accurate quantity information allows for earlier decisions on design changes and budget measures. The short turnaround also means repeat surveys are easy: if needed, you can re‑shoot the site to compare multiple development scenarios before construction starts. What used to be a one‑time survey due to time and cost constraints can now be repeated whenever conditions change.

Handling Sites Difficult to Survey (Slopes and Inaccessible Areas)

The benefits of drone surveying are especially large for sites that are dangerous or difficult for people to enter. On steep slopes or areas prone to collapse, ground surveying traditionally required personnel with safety harnesses, carrying high risks of omission and human error. With drones, you can obtain detailed data of hazardous spots by flying the aircraft from a safe distance. Targets that were difficult to measure in situ, such as scree slopes or quarry faces, can now have their shapes captured by aerial photos and converted into point clouds.

Drones are also highly useful immediately after disasters or during inspections of aging infrastructure. For instance, following a large landslide, access may be restricted due to aftershocks or secondary hazards, but a drone can scan the entire affected area from above. This enables early approximate quantity estimates (e.g., volume of collapsed material or emergency fill requirements), accelerating initial response.

Furthermore, inspections and measurements of high structures such as bridges and dams increasingly rely on drones rather than scaffolding or aerial work platforms. For locations that are difficult for a drone camera to capture—underwater surfaces or under forest canopies—there is growing use of ground‑based laser scanners or UAV LiDAR (LiDAR‑equipped drones) in combination. In this way, drone and 3D measurement technologies make previously unmeasurable sites measurable, dramatically improving on‑site safety and data accuracy.

Further Improving Site Coordinate Accuracy by Combining with LRTK

While drone photogrammetry produces point cloud models with very high relative shape accuracy, they may not inherently guarantee precise absolute coordinates. To align a point cloud to local coordinate systems (e.g., plane rectangular coordinates or public coordinate systems), a number of control points (GCPs: Ground Control Points) must be placed and their high‑precision positions provided. LRTK is powerful for surveying these control points.

LRTK (※ a smartphone RTK solution provided by Reflexia) uses a tiny RTK‑GNSS receiver that attaches to a smartphone, enabling centimeter‑level positioning easily. *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 specialized equipment and skilled operators, but LRTK has made the era of one smartphone per person for high‑precision surveying a reality. By combining a smartphone and a compact device and launching a dedicated app, you can obtain real‑time coordinates of any point with the press of a button.

If you provide GCPs measured with LRTK to the drone‑derived point cloud, the entire point cloud can be aligned to public coordinate systems with high accuracy. This allows the resulting 3D data to be overlaid precisely with design drawings and other survey data. LRTK is useful not only for single‑point surveying, but also for acquiring point clouds with absolute coordinates by leveraging the smartphone’s camera and AR features. For example, by simply walking the site with a smartphone you can obtain point clouds with absolute coordinates—useful for surveying indoor spaces or the undersides of bridges where drones cannot fly. Combining drones with LRTK enables high‑accuracy coverage of every point on site, further improving the data accuracy needed for approximate quantity design.

Conclusion: 3D Measurement Changes Estimate Accuracy and Workstyles

As described above, drone‑based 3D measurement is attracting attention as a technology that dramatically increases the accuracy and efficiency of on‑site approximate quantity design. Earthwork calculations that formerly required days can now be performed almost in real time, and dangerous surveying tasks are being replaced by safer methods. Consequently, accurate quantity information can be obtained from the design stage, enabling rational planning based on data rather than conservative safety margins or heuristics. This benefits both clients and contractors by optimizing budgets, reducing contract change risks, and even lowering environmental impact (by reducing unnecessary excavation and transport).

There are also changes in workstyles. Reducing the burden of surveying lets engineers produce results in less time and use the freed time for design studies and other work. Carrying heavy survey equipment and walking dangerous sites is becoming less necessary; instead, desk‑based analysis and utilization of point cloud data are becoming widespread. For young engineers, work that leverages digital tools is attractive, aiding recruitment and training. In short, the digital transformation (DX) brought by 3D measurement is simultaneously improving estimate accuracy and workstyle reform in approximate quantity design.

With the further promotion of i‑Construction policies by the Ministry of Land, Infrastructure, Transport and Tourism, the use of such 3D surveying technologies is expected to become even more widespread. If these methods become standard practice on sites, it will be possible to manage projects with digital data consistently from the approximate stage through completion, dramatically improving construction productivity and transparency.

Appendix: A Smartphone × High‑Precision Survey Workflow to Start with LRTK

Finally, we touch on a workflow for smartphone × high‑precision surveying—another approach gaining attention alongside drone surveying. Using the aforementioned LRTK, an ordinary smartphone can quickly become a high‑precision surveying tool. The basic on‑site workflow is simple.

• Attach device and start: Attach the LRTK Phone receiver to your smartphone and launch the dedicated app. After initial setup and receiving augmentation signals (such as the Japanese GPS satellite "Michibiki" CLAS signal), you are ready.

• Point measurement: Tap the button on the smartphone screen at the point you want to measure to record the coordinate (latitude, longitude, elevation). Measurement is instantaneous, and accuracy is typically within a few centimeters horizontally and vertically depending on conditions. You can also save point names and notes.

• Cloud sharing: Recorded data can be uploaded to the cloud on the spot. There is no need to return to the office for cable connections or manual data整理. On the cloud you can view measured points on a map and share data among multiple users for real‑time use.

This kind of simple surveying with LRTK is vastly more convenient and faster than traditional equipment. For example, leveling surveys that previously required multiple personnel can in some cases be replaced by a single LRTK‑equipped smartphone. Its mobility makes it a "daily‑use survey tool" that allows immediate measurements when needed, transforming on‑site workflows. Combining drone‑based wide‑area 3D measurement with pinpoint high‑precision positioning by a smartphone is an unbeatable combination. If you are about to undertake approximate quantity design, consider actively adopting these latest tools. A new workflow that balances efficiency and accuracy will strongly support your project.