How to Automatically Assign Coordinates to LiDAR Data｜5 Tips to Improve Accuracy

Point cloud data acquisition using LiDAR is becoming an indispensable technology for surveying in construction and civil engineering and for inspecting structures. However, the process of aligning the acquired point cloud data with an actual coordinate system is time-consuming, and ensuring accuracy remains a challenge. Many field engineers are likely interested in "methods to automatically assign coordinates to LiDAR data." If coordinates could be automatically assigned to each point in the point cloud, data processing efficiency would be greatly improved and the acquired 3D data could be put to immediate use.

This article explains a method to automatically assign coordinates to point cloud data obtained by LiDAR. It then introduces five tips to improve positioning accuracy. By acquiring high-accuracy coordinate-enabled point cloud data, you can smoothly utilize measurement data in various field operations such as as-built management, quantity measurement, and maintenance management. Please use this as a reference to achieve efficient and accurate 3D measurement.

Table of Contents

• What is automatic coordinate assignment to LiDAR data

• How to automatically assign coordinates to LiDAR data

• Tips for improving accuracy 1: Use high-precision positioning systems

• Tips for improving accuracy 2: Perform thorough pre-survey equipment calibration and planning

• Tips for improving accuracy 3: Optimize the positioning environment and perform stable measurements

• Tips for improving accuracy 4: Utilize control points and perform data validation

• Tips for improving accuracy 5: Ensure proper coordinate transformations and data processing

• Summary

What is automatic assignment of coordinates to LiDAR data

Automatic assignment of coordinates to LiDAR data refers to assigning accurate position coordinates (X, Y, Z coordinates) in real time to each measurement point of point cloud data acquired by laser scanners or LiDAR sensors. Typically, point cloud data are recorded in the measurement device's local coordinate system, that is, relative coordinates with the scanner itself as the origin. Therefore, to use raw point clouds overlaid on maps or design drawings, it is necessary after acquisition to match reference points and perform coordinate transformations (georeferencing) to align them with real-world coordinate systems.

Point clouds without coordinates required time-consuming and labor-intensive post-processing alignment, and they posed efficiency challenges because the placement of targets (control points) and additional surveying work were necessary. If coordinates can be automatically assigned, each point will have accurate map coordinates at the moment of capture. For example, "point cloud data with coordinates" generated by LiDAR scans in the field contains numeric values in a survey coordinate system or a public coordinate system for each point, so it can be handled in alignment with other GIS data and CAD drawings from the moment of acquisition. Traditionally, point cloud data had to be matched to known control points in post-processing, but a point cloud with automatically assigned coordinates has the major advantage of being usable immediately without additional alignment work.

How to automatically assign coordinates to LiDAR data

In LiDAR measurements, to automatically assign coordinates it is necessary to perform high-precision self-positioning simultaneously with the measurement and calculate three-dimensional coordinates for each point. As a specific method, it is common to use positioning techniques that combine GNSS (Global Navigation Satellite System) and inertial measurement units (IMU). For example, for drone-mounted LiDAR the aircraft is equipped with a high-precision GNSS receiver and an IMU, and by performing laser scanning while positioning the platform in real time during flight, the acquired point cloud can be assigned geographic coordinates on the spot. Similarly, in mobile mapping systems (vehicle-mounted LiDAR), GNSS/IMU track the vehicle’s trajectory and assign absolute coordinates to the point cloud.

In recent years, even mobile-device LiDAR (LiDAR sensors built into smartphones and tablets) has become capable of automatic coordinate tagging by combining it with high-precision GNSS. The positioning functions built into smartphones normally have errors on the order of several meters (several ft), but if centimeter-level positioning (cm level positioning; half-inch accuracy) is performed using an external RTK-GNSS receiver or the like, it is possible to assign highly accurate coordinates to each point during scanning with a smartphone. In this way, by linking LiDAR sensors with high-precision positioning devices, a system can be realized that automatically assigns coordinates at the moment point cloud data are acquired.

As a conventional method, targets or control points with known coordinates are installed on-site in advance, and after scanning they are matched with the point cloud to assign coordinates. Although this is not fully automated, georeferencing can be performed semi-automatically by matching corresponding points in software. However, this approach requires additional control-point surveying and target installation effort, and can be difficult under certain site conditions. Therefore, a system that incorporates GNSS real-time positioning to directly assign coordinates to the point cloud remains superior in terms of efficiency and immediacy.

Tip 1 for Improving Accuracy: Use High-Precision Positioning Systems

When automatically assigning coordinates, ensuring the accuracy of the positioning system itself is most important. Use high-precision GNSS positioning technology to minimize the positional error of each point as much as possible. Specifically, by using GNSS receivers that support error-correction technologies such as real-time kinematic (RTK) and post-process kinematic (PPK) methods, you can achieve position accuracy on the order of several cm (a few in). This makes it possible to assign coordinates to LiDAR point clouds with cm-level accuracy (half-inch accuracy). By contrast, standard standalone positioning (standalone GPS) can produce errors on the order of several m (several ft), making it difficult to accurately align point cloud data with maps.

When using high-precision GNSS, selecting compatible equipment and services is also important. There are methods such as installing a base station to perform relative positioning with a rover, and methods of receiving augmentation signals from satellites or ground-based infrastructure (using SBAS or cloud-based correction services). In any case, a positioning system that supports RTK/PPK is indispensable to obtain centimeter-level (cm level accuracy (half-inch accuracy)) accuracy. Also, not only the GNSS receiver but the accuracy of the IMU built into the platform or equipment is important. The IMU measures the platform’s attitude and motion and plays a role in estimating position when GNSS is temporarily interrupted. Systems equipped with a high-quality IMU can maintain highly accurate position estimates even during brief GNSS signal interruptions, which in turn leads to improved accuracy of point cloud coordinates.

Tip 2 for Improving Accuracy: Thorough Pre-Calibration of Equipment and Survey Planning

Before conducting LiDAR surveys in the field, calibrating the equipment to be used and preparing a thorough surveying plan are also fundamental to ensuring accuracy. First, accurately measure the mounting position relationship (lever arm) between the LiDAR sensor and the GNSS antenna and IMU, and enter it into the system as needed. By thoroughly performing calibration work to correct misalignments between instruments, you can prevent systematic errors from entering the acquired data. Especially for mobile mapping and drone LiDAR, it is essential to appropriately correct misalignments between the coordinate systems of the laser scanner and the GNSS/IMU (boresight angles and position offsets).

Also, during the survey planning phase, carefully consider the route, altitude, and attitude for taking measurements. For drones, plan the flight altitude and flight course; for ground-based surveys, plan the scanning area and placement locations—making these plans in advance can lead to uniform data collection and stable positioning. In addition, check the GNSS reception status and satellite configuration (satellite geometry) before measurement. Because the satellite configuration can be biased at certain times of day and cause decreased positioning accuracy, adjusting the survey date and time can also be effective. By arriving on site with thorough preparation, you can maximize the accuracy of automatic coordinate assignment.

Before the actual measurements, it is advisable to perform test measurements on targets with known distances or heights (for example, a calibration scale of a specified length or benchmarks) to verify that the system is measuring accurately. A simple preliminary check can prevent unexpected equipment failures or the occurrence of errors.

Tip 3 for Improving Accuracy: Optimize the Positioning Environment and Ensure Stable Measurements

To maintain high accuracy during field measurements, it is important to optimize the positioning environment and always strive for stable data acquisition. For GNSS positioning, ensuring a good line of sight to the positioning satellites is fundamental. In areas where buildings or trees are densely clustered, satellite signals can be blocked or reflected, which may greatly degrade positional accuracy. For example, on sites in urban areas with many high-rise buildings, signal reception at ground level surrounded by buildings becomes extremely unstable. In such cases, measures such as combining GNSS observations from building rooftops or conducting measurements during periods of favorable satellite geometry are effective. Also, when possible start measurements from an open area and avoid long-duration positioning in locations with poor radio conditions, such as urban canyons between tall structures or deep inside forests. If surveying must be carried out under site conditions with many obstacles, it is effective to use positioning services that are more readily received from above (satellite augmentation), or to break measurements into short intervals while moving to sequentially recover GNSS reception.

Meanwhile, attention is also needed to the stability of the LiDAR scan itself. If the LiDAR is mounted on a moving platform, avoid rapid acceleration and sudden stops, and maintain as constant a speed and attitude as possible while driving or flying to stabilize IMU-based corrections. When scanning with a handheld device (such as a smartphone), move the device as smoothly as possible and be mindful not to change its orientation abruptly. If GNSS reception deteriorates during measurement, you may need to stop measuring, move to a location where positioning becomes stable, and then resume. Always monitor the device status and reception of correction information, and optimize the environment and measurement methods to obtain high-quality data.

Tip 4 for Improving Accuracy: Utilize Reference Points and Perform Data Validation

Even when point cloud data are georeferenced using an automatic coordinate-assignment system, reliability can be further increased by installing known reference points (control points) or checkpoints on site and performing verification and correction of the results. Georeferencing with high-precision GNSS theoretically yields high accuracy, but in actual surveying small satellite positioning errors and sensor drift can cause displacements on the order of several centimeters (several in) to over 10 cm (over 3.9 in). To correct this, multiple reference points are established within the survey area and their coordinates are measured precisely with conventional surveying instruments (total stations, etc.). If the acquired point cloud data include points corresponding to those reference points, software can be used to match them and calculate the errors, enabling correction of the overall coordinates. Note that reference points should be placed appropriately, including around the perimeter of the survey area, and performing corrections and verifications at multiple locations prevents bias in accuracy.

Similarly, to evaluate the accuracy of point cloud data, it is also recommended to check errors at several checkpoints (validation points) after measurement. For example, use corners of important structures or known elevation points as checkpoints and compare the coordinate values on the point cloud with the true values that were surveyed beforehand. This allows you to determine whether there are systematic errors in the automatic coordinate assignment results and whether the accuracy level meets the required standards. If a large deviation is confirmed, consider applying a global translation and rotation correction to the entire dataset within the software. By combining correction and validation using control points, the accuracy assurance of the final point cloud deliverables becomes more reliable.

Tip 5 for Improving Accuracy: Ensure Proper Coordinate Transformations and Thorough Data Processing

When ultimately using acquired georeferenced point cloud data, it is also important to properly handle coordinate systems and data processing. Positioning using GNSS yields location information in geodetic latitude and longitude (or ellipsoidal height) in the World Geodetic System, but this may differ from the coordinate systems used in practice (for example, Japan’s plane rectangular coordinate system or elevations). Therefore, perform geodetic datum transformations and height conversions (geoid height conversion) accurately when necessary. For example, heights obtained from GNSS are ellipsoidal heights and have a different reference from Japanese elevations. By converting heights using the Geoid Model (geoid height) of the Geospatial Information Authority of Japan, you can align the height information of point cloud data with the reference benchmarks. Also, when performing coordinate transformations, mistakes in the geodetic datum or surveying parameters can shift positions by tens of cm (several in) to several m (several ft), so it is important to thoroughly check the settings of the software you use.

Also, quality control of the point cloud data itself is essential. If the data contain noise points or outliers, remove them by filtering, and apply smoothing or thinning (downsampling) as needed. When merging point clouds acquired in multiple passes, verify that each dataset is aligned to the same coordinate system, and if there are discrepancies, reproject them to a unified coordinate system. By thoroughly performing these tasks during the data processing stage, you can prevent degradation of point cloud accuracy and produce reliable surveying deliverables.

Summary

Automatically assigning coordinates to LiDAR data dramatically improves the efficiency and utility of point cloud surveying. Because three-dimensional data acquired at survey sites can be immediately matched with maps and drawings, it can be quickly applied to various uses such as as-built verification, progress management, and design comparisons and validations. This article introduced the method and key points for improving accuracy. The important thing is to combine high-precision positioning technologies with careful planning and verification to obtain stable georeferenced point clouds.

Furthermore, in recent years high-precision GNSS receiver devices that can be attached to smartphones (e.g., LRTK) have also appeared. By combining such devices with a smartphone’s built-in LiDAR scanner, it is possible to easily scan the surroundings on-site while assigning centimeter-level coordinates to each point in real time (cm level accuracy; half-inch accuracy). For example, even for small-scale on-site as-built verification or simple surveying, an era has arrived in which anyone can perform highly accurate 3D measurements in a short time with just a smartphone and a high-precision GNSS. The assignment of coordinates to point clouds, which traditionally required specialized equipment and advanced manual work, is being greatly simplified. Let’s adopt new technologies to bring efficient and high-precision 3D surveying to the field. Such efforts will further accelerate DX (digital transformation) in the construction and civil engineering sectors.