How to Create High-Precision 3D Models by Combining RTK and Point Cloud Surveying

This article takes an average of 2 minutes and 30 seconds to read

Published February 28, 2025

Basics of RTK Positioning and Point Cloud Data

​

What is RTK?

RTK-GNSS (Real-Time Kinematic Positioning) is a positioning method that provides high-precision location information by simultaneously observing two locations: a reference station with a known position and a mobile station, which is the object being measured. By transmitting real-time satellite positioning error correction data from the reference station to the mobile station, the positioning accuracy is greatly improved compared to standalone positioning. Specifically, the accuracy is enhanced to approximately 2-3 cm horizontally and 3-4 cm vertically, making it widely used in fields that require centimeter-level precision, such as civil surveying. RTK generally sends correction data to the mobile station via radio or network (such as NTRIP), allowing high-precision global coordinates to be instantly obtained anywhere on-site.

​

What is Point Cloud Surveying and 3D Point Cloud Data?

Point cloud data is a collection of countless measurement points (points) that make up the surface of objects or terrain. Each point contains "X, Y, Z" three-dimensional coordinates, and when combined with color images, it is possible to assign color (R, G, B) information to each point. The collection of points visually represents the shape of the object in great detail, much like a photograph, and since it also includes height information (Z values), it can be visualized in three dimensions. For example, point cloud data obtained from the latest LiDAR scanner for an urban intersection captures the intricate details of buildings and roads. This 3D point cloud surveying technology is gaining attention in the civil engineering industry because it can record complex current shapes that are difficult to understand with traditional 2D drawings.

There are various methods for acquiring point cloud data. Generally, the following approaches are used for 3D surveying to obtain point clouds:

• Ground-based 3D Laser Scanning (TLS) – Using a tripod-mounted laser measurement device to scan buildings and terrain with high precision and density.

• UAV (Drone)-mounted Laser Surveying – Attaching a LiDAR sensor to a drone to quickly capture a wide range of point clouds from the air.

• Mobile Mapping Systems (MMS) – Mounting a laser scanner on a vehicle or other mobile platform to collect point clouds while moving through the surrounding area.

• Photogrammetry (SfM) – A method that analyzes multiple photos taken by drones or DSLR cameras to reconstruct 3D shapes from matching key points and generate point clouds.

​​

These methods allow for the acquisition of point cloud data consisting of millions to billions of points. However, the raw point cloud data immediately after acquisition may contain unwanted points or noise, so processing with dedicated software and alignment of multiple datasets is required, as discussed later. The completed point cloud data is used in various civil engineering applications, such as creating 3D models of current terrain, generating cross-sectional drawings, and verifying discrepancies with design data.

​

Benefits of Combining RTK and Point Cloud Surveying

​

By combining high-precision positioning with RTK and point cloud surveying, highly accurate 3D models can be efficiently created. One of the biggest advantages is the ability to instantly assign global coordinates (world coordinates) to the acquired point cloud data. For example, when point clouds are captured using RTK-compatible surveying equipment, all the points captured can automatically be assigned coordinates with centimeter-level accuracy. This significantly simplifies the alignment process in post-processing, and allows for immediate on-site measurement of as-built conditions and dimension measurements.

Moreover, since RTK constantly corrects the positioning in real-time, it prevents point cloud distortion or drift (cumulative errors) that can occur when walking and scanning with smartphones or handheld scanners. In fact, with typical smartphone LiDAR measurements, long-distance walking often results in ground distortion, but by integrating RTK, the position during scanning is continuously corrected, reducing error accumulation. Additionally, using RTK eliminates the need to set up reference points or target markers, which is another major benefit. Traditionally, to align point cloud data with the correct coordinate system, known points had to be set up on-site and measured, but with RTK, coordinates can be directly assigned to the measurement data without the cumbersome setup of reference points.

In this way, the fusion of "high-precision positioning" with "high-density shape information" enables the creation of 3D models that exceed traditional methods in terms of accuracy, efficiency, and convenience.

Use Cases of RTK × Point Cloud Data

​

High-precision 3D models with point clouds can be utilized in various applications on civil engineering sites. Here, we introduce four specific use cases that are of particular interest to target groups in the civil engineering industry.

​

Application in As-Built Management

​

By converting structures and terrain into point cloud data after construction, the process of as-built management (as-built inspection) is significantly streamlined. Traditionally, completed embankments and structures were measured point by point and compared with design values. However, by 3D scanning the entire site with point cloud data, a comprehensive comparison with the design's 3D data becomes possible. In practice, overlaying the acquired point cloud with the 3D design data allows for visual assessment of the as-built condition, as well as simultaneous calculations of soil volume and verification of progress.

The Ministry of Land, Infrastructure, Transport, and Tourism has also developed guidelines for as-built management using 3D measurement technologies, and methods using RTK-GNSS, TLS, and UAV photogrammetry are being promoted for as-built management. By utilizing point cloud data, the entire process—from on-site as-built measurements to the creation of inspection reports—can be simplified, improving both the accuracy of quality control and operational efficiency.

​

Soil Volume Calculation Using Point Cloud Data

​

Point cloud data significantly enhances the accuracy of soil volume calculations (volume estimation) in earthworks. For example, by measuring the terrain before and after excavation using drones or laser scanners, and calculating the fill and excavation volumes from the two generated surface models, a more detailed and reliable estimate of the soil volume can be obtained compared to traditional cross-sectional methods. Since point cloud data reflects the terrain in great detail, subtle changes in slopes and small depressions caused by the burial of pipes can also be incorporated into the soil volume calculations. Additionally, by comparing the design model with the current point cloud data before construction, the required amount of embankment can also be estimated.

Major general contractors are already implementing practices where point clouds and 3D design data are compared to calculate soil volumes and verify progress, proving the high accuracy and efficiency of these methods. Point cloud-based soil volume measurement significantly reduces manual labor required for soil quantity takeoffs and contributes to improving the fairness of progress and quantity management.

​

Application in Infrastructure Maintenance

​

RTK × point cloud technology is also beginning to play a significant role in the maintenance of public infrastructure such as bridges, roads, and tunnels. With the increasing shortage of personnel for aging infrastructure inspections, point cloud data allows for detailed documentation and analysis of structures with fewer people and in less time. For example, in the inspection of small bridges, workers can hold an iPhone and scan the bridge’s piers and beams using the built-in LiDAR sensor, capturing the point cloud data by themselves. By attaching a network RTK-GNSS receiver to the smartphone, the positioning accuracy is enhanced, and accurate coordinates are assigned to the acquired bridge point cloud data.

The resulting 3D point cloud model can store damage photos and notes about deteriorated areas, digitally linking them as opposed to relying on handwritten records in a field notebook. Additionally, the smartphone can measure the length of cracks, the area of the repair range, or the volume of accumulated soil, reducing the need for measuring tapes or other tools.

This use of RTK-enabled devices and point cloud surveying for infrastructure inspections allows for the precise storage of the current condition of structures as 3D data, which is useful for future comparisons of changes or repair planning. In tunnel and road maintenance, point clouds captured by laser scanning are being used to monitor displacement and detect abnormalities early. As this technology advances, it is expected to become an indispensable tool for the future development of infrastructure management.

​

Application in Construction Simulation

​

The completed high-precision 3D models are also useful for construction simulations and pre-implementation verification. By overlaying the design models of structures on the point cloud data of the current terrain, potential interference points and constructability can be identified before construction begins. For example, heavy equipment models, such as those for bridge inspection vehicles, can be placed on the acquired 3D point cloud to simulate and verify if the actual work can be performed without issues.

In a case with the Tokyo Metropolitan Expressway Company, the movement of high-altitude work vehicles was simulated on point cloud data, allowing for optimal vehicle selection, review of work procedures, and checks for interference with obstacles, which significantly improved on-site work efficiency. Similarly, when planning construction procedures, the routes for heavy equipment transportation or excavation areas can be simulated on the point cloud model to verify if work in narrow spaces is possible and ensure that temporary structures do not interfere with the terrain.

Additionally, by combining AR (augmented reality) technology, the point cloud data and design models can be overlaid and displayed through a tablet on-site. This allows workers to intuitively visualize the completed project and take preventative measures to avoid construction errors.

In this way, by utilizing 3D models beforehand, the need for “trial and error” during actual construction is reduced, helping to prevent potential construction issues.

​

Specific Implementation Methods and Key Considerations

​

Next, we will explain the specific methods and key considerations for implementing RTK and point cloud surveying on-site. Keeping the entire workflow from data acquisition to processing and utilization in mind, let's focus on the points for effective implementation.

​

Methods of Acquiring Point Cloud Data

​

As mentioned earlier, there are two primary methods for acquiring point cloud data: laser scanning and photogrammetry.

For laser scanning, you need to choose from various types of equipment, such as ground-based stationary, mobile, or UAV-mounted laser scanners. For example, if you want to record detailed features of a structure with high precision, a stationary ground laser scanner is used. For surveying large areas of terrain, UAV-mounted laser scanners or photogrammetry are more suitable. In the case of photogrammetry, multiple photos are taken with a drone or a DSLR camera, and these images are processed using specialized software and Structure from Motion (SfM) technology to generate point cloud data.

In recent years, LiDAR sensors built into devices like iPhones and iPads have made it easy to acquire point clouds within a few meters of range, making smartphone surveying a viable option for smaller sites.

By combining RTK positioning with any of these methods, high-precision position data can be obtained simultaneously with measurements, ensuring smoother post-processing. For instance, using an RTK-enabled drone allows for high-precision point clouds to be acquired without needing to place numerous ground control points (GCPs).

Depending on the site scale, required accuracy, and budget, select the optimal measurement method and equipment.

​

Methods of Integrating RTK and Point Cloud Data

​

There are two main methods for data integration (georeferencing): directly geotagging with RTK and aligning with reference points during post-processing.

In the first method, when using RTK-enabled drones or surveying equipment, high-precision coordinates can be instantly assigned to the acquired data (photos or point clouds). For example, when using an RTK-equipped drone for photogrammetry, the position coordinates of the photos themselves will be high-precision, and during analysis, the point cloud model will automatically align with the correct coordinate system.

In the second method, even if using post-processing, known reference points set up on-site can be surveyed with RTK or other methods, and the point cloud can be aligned to these coordinates afterward. In ground-based laser scanning, multiple target markers (such as spherical markers) are placed on-site, and their positions are measured using a total station or RTK. The target points included in the point cloud are then linked and the entire data set is transformed. In photogrammetry, by preparing reference points with known coordinates (such as ground markers) and marking them on the photos, the model can be aligned to geodetic coordinates.

​

The key advantage of using RTK is that it significantly reduces the effort required for coordinate alignment, which is especially beneficial for large-scale surveys, leading to substantial time savings and improved accuracy. During data integration, it's important to convert to the required surveying coordinate systems (such as plane rectangular coordinates) or adjust for geoid height differences, ensuring consistency with other surveying results.

​

Selecting Software for Analysis and Utilization
 

When processing and analyzing acquired point cloud data, selecting the right software based on the application is crucial. Various specialized tools are available for point cloud processing, such as noise removal, merging point clouds, and coordinate transformation. For example, software like FARO's "Scene" and Trimble's "RealWorks" are commonly used to load point cloud data from laser scans, delete unnecessary points, and merge data. These programs often include automatic noise and outlier filtering features, but manual cleaning may still be necessary for high-precision results.

For photogrammetry, software like Pix4Dmapper, Agisoft Metashape, and the domestic WingEarth handle the process of generating point clouds from photographs. When analyzing point cloud data, free software like CloudCompare or commercial software like Autodesk Recap can be used to visualize the data and perform tasks such as cross-section extraction, shape comparison, or volume calculation. Recently, point cloud viewers and analysis services that operate in the cloud have emerged, allowing point cloud visualization and basic measurements directly in a browser without the need to install specialized software.

For final 3D modeling (mesh creation), tools that convert point clouds into polygon models, such as Bentley ContextCapture, Geomagic, and ClassNK-Peerless, are used. Among these software options, choose the one that best matches your company's specific needs (e.g., whether you want to calculate soil volumes or convert to design CAD data) and the skill level of the operator. Additionally, when implementing the software, it's important to consider compatibility with existing systems and standardizing data formats (such as LAS, PLY, E57, etc.).

​

Utilization and Benefits of LRTK

​

Finally, we summarize the features and benefits of LRTK, a solution that makes RTK × point cloud technology easily accessible. LRTK is a product line developed by Lefixia Inc., a startup originating from the Tokyo Institute of Technology. It combines smartphones and dedicated LiDAR devices with RTK-GNSS to enable anyone to perform high-precision surveying with ease. In this section, we will introduce LRTK Phone and LRTK LiDAR as two key devices, as well as the cloud integration service, and provide examples of how they can be utilized in civil engineering projects.

​

LRTK Phone: Easy Point Cloud Scanning with iPhone + RTK

​

LRTK Phone is a pocket-sized RTK-GNSS receiver that integrates with iPhones and iPads. This ultra-compact device attaches to the back of the smartphone and performs centimeter-level positioning while receiving corrections through network RTK. When connected with the dedicated "LRTK" app, the smartphone transforms into a high-precision, all-purpose surveying tool. Specifically, it can handle tasks like positioning (point measurement), 3D point cloud scanning, layout marking (positioning), and even AR simulations—all with a single device. The acquired data can be instantly shared on the cloud.

While traditional surveying required separate devices for these functions, LRTK Phone combines them all in one lightweight device, weighing only about 125g, making it easy to carry. Its price is also set more reasonably than traditional surveying equipment, making it a practical solution for "one device per person" deployment.

​

LRTK Phone Device (black body section) attaches to the smartphone. This device enables the smartphone to perform RTK positioning, providing centimeter-level position accuracy.

​

The standout feature of LRTK Phone is its ability to integrate the built-in LiDAR scanning function of the iPhone/iPad with RTK positioning, enabling point cloud scanning with high-precision coordinates. Although iPhone/iPad LiDAR enables easy 3D point cloud measurement, the usual positioning accuracy is low (around GPS accuracy of a few meters), and there was a challenge where position errors gradually accumulate during scanning, distorting the terrain. However, with LRTK Phone, RTK allows the current position to be continuously tracked with centimeter-level accuracy, so all the acquired point clouds are assigned global coordinates, and the data is corrected to prevent distortion even during walking scans.

The design is intuitive, requiring no complicated operations, making it easy for on-site workers to use. With just the smartphone and LRTK device, point cloud scanning can be quickly performed on-site, and after scanning, it’s possible to instantly measure distances between any two points, areas, or volumes. For example, checking the cross-sectional shape of an embankment on-site or immediately calculating the volume of an excavation area can now be done using just the smartphone, without needing heavy equipment or a computer.

In this way, LRTK Phone is a powerful tool that supports instant measurement and immediate data utilization on-site.

​

LRTK LiDAR:

High-Precision Measurement with RTK + Long-Range LiDAR

​

LRTK LiDAR is a long-range 3D measurement device that combines GNSS-RTK technology with high-resolution laser scanning technology. It is positioned as a "marker-free" revolutionary 3D measurement system that automatically assigns centimeter-level coordinates to the point cloud data during scanning, without the need for markers or known reference points. Traditionally, laser scanning over large areas required placing targets on-site or performing alignment during post-processing, but with LRTK LiDAR, these tasks are eliminated while still obtaining high-precision point clouds.

​

A notable feature is its measurement range, as it can scan and generate point clouds with high precision for structures up to 200 meters away. While typical LiDAR on mobile devices can measure about 5 meters, and ground-based laser scanners typically measure from several dozen to several hundred meters, LRTK LiDAR is capable of surveying large civil engineering sites and large-scale structures. The vast point cloud data obtained can also be quickly previewed on a smartphone in just a few minutes, allowing for flexible operations where the scanning results can be checked on-site and additional scans can be performed immediately if necessary.

Designed to allow anyone to easily and accurately acquire high-precision point clouds without errors, LRTK LiDAR is also user-friendly, even for those who are not professional surveyors. It is well-suited for complex terrains or intricate sites like plant facilities and pipelines. The acquired point cloud data can directly contribute to creating 3D models for civil CIM and BIM, as well as advanced analysis for infrastructure inspection data.

​

Cloud Integration and Utilization of Point Cloud Data

​

Point cloud data acquired with the LRTK series becomes even more accessible and easier to utilize when integrated with the dedicated LRTK cloud service. By uploading the point cloud data scanned on-site to the cloud, it can be instantly viewed and shared through a web browser from a PC in the office. For example, with the point cloud viewer feature on the LRTK cloud, the uploaded point cloud can be displayed in 3D in the browser, and users can check the coordinates of each point, as well as measure distances, areas, and volumes. Since there is no need to install specialized software, sharing data with stakeholders at remote locations is easy, allowing for real-time monitoring and discussions of the situation.

​

On the cloud, it is possible to overlay point cloud data with coordinates and design 3D data to check as-built conditions, as well as extract the necessary cross-sections for creating drawings, even without a PC with the corresponding software installed. In terms of security, managing everything on the cloud makes it easier to track data updates and manage access permissions, allowing for better control over information sharing. The LRTK cloud serves as a hub that connects "on-site ⇔ office", supporting the effective utilization of surveying data.

​

As shown above, the creation of high-precision 3D models by combining RTK and point cloud surveying brings significant benefits across various aspects of civil engineering. Especially with solutions like LRTK Phone and LRTK LiDAR, we are approaching an era where anyone, even without specialized knowledge, can easily acquire and utilize high-precision point cloud data. If you're interested, please check the Lefixia Inc. website for product details and implementation examples. Currently, the official site also offers free documentation requests regarding LRTK, allowing you to obtain more detailed information.

​

We encourage you to incorporate the latest technology for creating high-precision 3D models on-site, and use it to enhance the productivity of surveying and construction management tasks.