6 Tips to Reduce Errors in Smartphone LiDAR Surveying | Check Before Using on Site

Smartphone LiDAR surveying is an effective method when you need to quickly grasp the shape of a space on site. Without bringing heavy dedicated equipment every time, you can capture the spatial situation while walking, so its use is expanding in various situations such as pre-construction checks, overview of as-built conditions, records before renovation, and pre-checks for interferences. At the same time, when practitioners actually start using it, it is also true that they often face problems such as point clouds being more distorted than expected, walls appearing wavy, floors tilting, and coordinates not matching.

To stabilize results from smartphone LiDAR surveying, simply walking with the device is not enough. Errors depend not only on the device itself but also on site conditions, walking method, the target objects, how references are taken, and how you check after acquisition—essentially the whole operation. In practice, even if the model looks plausible, problems that arise later include: it does not align correctly with the necessary positions, measurements used for dimension checks are off, or datasets captured on different days do not align. Here, from a practical perspective, we explain in detail six tips you should confirm before using smartphone LiDAR surveying on site to reduce errors.

Table of contents

‐ Reasons why errors become large in smartphone LiDAR surveying ‐ Tip 1: Decide the measurement purpose and required accuracy first ‐ Tip 2: Check site conditions before measurement ‐ Tip 3: Stabilize your walking and device orientation ‐ Tip 4: Reduce misalignment with control points and known points ‐ Tip 5: Check the point cloud immediately after acquisition and re-capture if necessary ‐ Tip 6: Separate correction and validation by deliverable ‐ Operational rules to confirm before using on site ‐ Summary

Reasons why errors become large in smartphone LiDAR surveying

To reduce errors in smartphone LiDAR surveying, it is important first to organize where errors originate. Many people tend to attribute errors solely to device performance, but in reality that is not the only factor. Smartphone LiDAR surveying simultaneously involves the mechanism for ranging, the mechanism for estimating self-position from images, and the mechanism for compensating orientation. Therefore, results change depending on the target’s reflective characteristics, ambient brightness, walking speed, maintaining a clear line of sight, and how the same place is presented.

Errors tend to become large particularly in places where monotone wall surfaces continue, outdoors under strong direct sunlight, locations with many metals or glass, narrow spaces that restrict movement, and areas with many steps or obstacles where device orientation cannot be stabilized. In such environments, distance information to targets can become unstable and self-position tracking can be disturbed, causing small accumulated drifts to appear as large distortions in the end. On site, the first tens of seconds may look fine, but as the measurement area expands the overall consistency can collapse. This is a typical failure frequently seen in smartphone LiDAR surveying.

Another important point is that there are shape errors and positional errors. Shape errors refer to issues in reproducing the object’s form itself, such as walls bulging, floors sagging, and corners rounding. Positional errors refer to the whole dataset being shifted when overlaid with known coordinates, drawings, or data captured on another day. What causes trouble on site is operating without distinguishing these two. Even if something looks good visually, if the coordinates don’t match it is hard to use for layout work; conversely, even if coordinates are matched, local distortions still leave doubts for dimension checks. Therefore, countermeasures for errors in smartphone LiDAR surveying must be organized after clarifying what you want to make accurate.

Also, smartphone LiDAR surveying is not omnipotent. While it enables quick spatial recording, situations that require high accuracy guarantees—such as official control point surveys or strict as-built management—may need complementary methods depending on the application. The important thing is not to assume that a smartphone necessarily has low accuracy, nor to believe a smartphone can do everything alone. By understanding how errors occur and adopting appropriate usage, it becomes a powerful tool to speed up on-site decision making. With that premise in mind, we will now look at concrete tips.

Tip 1: Decide the measurement purpose and required accuracy first

The first tip is to decide in advance why you are measuring and what level of accuracy is required before starting smartphone LiDAR surveying. If this is left vague when entering the site, you may capture a wider area than necessary and increase distortion, or conversely fail to capture enough information in the areas you want to check and have to re-capture. In practice, the necessary approach changes completely depending on the purpose—overview of current conditions, pre/post-construction comparison, interference checks, renovation planning, rough quantity estimates, base for drafting, progress records, etc.

For example, if you want to quickly understand the existing conditions before equipment replacement indoors, what matters is the overall connectivity and relative positions of obstacles. In this case, rather than trying to capture a wide area at high density in one go, it is better to divide the necessary space and prioritize areas you do not want to miss; this reduces error. Conversely, if you want to examine floor irregularities or the positions of openings relatively finely, you need to stabilize the walking route, viewpoint height, and distance to the target. Different objectives change which error points matter.

One point to be aware of is that smartphone LiDAR surveying tends to be disadvantaged when you try to complete a wide area in a single session. On site, people often think, “Since we’re already there, let’s capture everything at once,” but that can increase distortion. Especially where indoor and outdoor areas are continuous, long corridors, or spaces with repetitive scenery, the longer the measurement time the more self-position estimation drift accumulates. For areas where required accuracy is not high, accept them as records, and perform careful, separate sessions for areas where accuracy matters—this kind of separation is effective.

It is also important to envision the deliverable at the outset. Whether you only need to glance at the point cloud, use it for sectional checks, use it as a base for floor plans, or overlay it with other data with coordinates changes the required operation. If you measure without envisioning the deliverable, data that seemed sufficient on site can turn out to be unusable in the office. The first step to reducing errors is, before measurement skill, to verbalize what and how accurately you want to capture. People who succeed with smartphone LiDAR surveying invest time in organizing these things before measuring.

Tip 2: Check site conditions before measurement

The second tip is to confirm site conditions before measurement and proactively eliminate factors that cause errors. In smartphone LiDAR surveying, if you only notice poor environmental conditions after starting acquisition, the quality will already be degraded. Upon arriving at the site, instead of starting immediately, first check lighting conditions, the presence of moving objects, highly reflective surfaces, entry restrictions, walking routes, and sightlines to the endpoint.

Outdoors, there are situations susceptible to the influence of direct sunlight. When strong light is present, how targets appear and tracking stability can change, and in places with large brightness contrasts data connectivity can deteriorate. Indoors, spaces that are too dark or areas near openings with strong backlighting also make stable tracking difficult. Changing the time of day slightly often makes acquisition easier. If you want to reduce errors, think of capturing under favorable conditions before worrying about device operation.

Be mindful of the material of target objects. Glass, mirrored surfaces, highly polished metal, monotone white walls, and floors with puddles are typical examples that destabilize shape recognition and distance measurement. In such places, even if walking looks fine, later inspection of the point cloud may reveal localized missing areas or waviness. Although it is difficult to avoid these on site entirely, stability can be improved by not using such surfaces as primary references, supplementing them from different angles, or intentionally including distinguishing visual markers in view.

Additionally, moving objects on site must not be overlooked. People and vehicles, swaying sheets, materials moved by wind, and opening/closing doors can disrupt continuity matching during continuous acquisition. If measurement overlaps with busy working hours, results may vary each time you measure. Smartphone LiDAR surveying is strong for static spaces but disadvantaged in environments with many dynamic changes. Choose times when people and machinery move less whenever possible; if unavoidable, separate the areas affected and capture them independently.

Pre-measurement checks are not meant to eliminate all errors but to avoid major failures. If you start acquisition without understanding site conditions, you will not know later what went wrong, leading to unreproducible operations. Conversely, by checking conditions first you can decide, for example, to revisit some outdoor areas at a different time, segment this section into shorter runs, or not use this wall as a reference. To use smartphone LiDAR surveying stably in practice, observation skills before measuring are as important as the measuring technique.

Tip 3: Stabilize your walking and device orientation

The third tip is to stabilize your walking and the device orientation. Errors in smartphone LiDAR surveying are greatly affected by how you move the device. On site you may be tempted to hurry and capture quickly, but moving too fast, changing direction frequently, sudden large shakes of the device, and abrupt changes in distance to the target all cause distortion. A common practical error is looking around too much while walking, which destabilizes tracking.

Basically, keep the device at a consistent height, avoid drastically closing in or moving far from the target, and proceed at a slow, steady speed. When people enter narrow spaces they unconsciously tilt the device or make sudden turns to avoid obstacles. These small disturbances accumulate over long paths. Pay particular attention at corners, entrances, near stairs, and in narrow passages where posture tends to break down. “Being careful” might make you think only of moving slowly, but the essence is regularity of movement. Maintaining a consistent rhythm and a similar view as you progress leads to more stable results.

It is also important to keep the device oriented so that the spatial features are sufficiently visible. If you keep looking only at the floor or focus excessively on walls, it becomes difficult to capture the overall spatial relationships. In smartphone LiDAR surveying, it is desirable to keep elements that aid tracking—floors, walls, corners, openings, fixtures, columns—moderately within the screen. When you track predominantly featureless surfaces, position confusion is likely where similar views repeat. In large spaces, simply avoiding continuously viewing monotone surfaces can make a difference.

Designing the measurement route is also important. If you walk randomly on site, even if you think you are looking at the same place, your viewpoints tend to become intermittent. Predefine the start and end points, places to turn back, areas you don’t want to miss, and convenient break points for re-capture; this helps stabilize movement. When looping, forcing yourself to close the loop quickly at the end can actually make drifts more noticeable. Rather than obsessing over closure, it’s better to proceed while maintaining quality along the way.

Although smartphone LiDAR surveying appears to be just walking with the device held up, it actually depends heavily on body operation. That is why results can differ by operator even with the same site and device. To reduce errors, standardizing on-site movement, not just device settings, is indispensable. Share basic rules for walking speed, device height, orientation, and actions at corners in advance so anyone can achieve comparable quality; doing so greatly reduces practical variability.

Tip 4: Reduce misalignment with control points and known points

The fourth tip is to use control points and known points to suppress positional misalignment. What is often overlooked in smartphone LiDAR surveying is that capturing shape well and being in the correct position are separate issues. Even when the three-dimensional data looks clean on site, overlaying it with known dimensions, existing drawings, or other positioning data may reveal an overall translation, rotation, or partial stretching/compression. To prevent this, it is important to have comparison references on site.

Control points do not have to involve elaborate operations. Simply deciding in advance points whose positions are known and can be rechecked on site, clear corners, points corresponding to grid lines, or points that can later be checked by other means is effective. Smartphone LiDAR surveying is strong in continuous acquisition, but if tracking fails at any point the subsequent positional relationships can shift in a chain reaction. That is why having points to cross-check mid-way or at the end is necessary.

Especially in practice, if you acquire without references it becomes difficult to evaluate results back in the office. Even if the point cloud looks good, you cannot judge if it is truly correct. Conversely, having known points makes it easier to understand in which direction and by how much the data is off, allowing you to separate usable and less usable areas. Reducing errors is often thought of as improving acquisition techniques, but in reality creating an evaluation yardstick on site is key to quality control.

Also, if you prioritize positional accuracy, it is important not to rely only on standalone smartphone acquisition. A practical approach is to use smartphone LiDAR to efficiently capture spatial shapes and complement positional referencing with another high-accuracy positioning method. With this mindset you avoid forcing all weaknesses into a single device and can more easily approach the overall accuracy required. The important thing is to leverage what smartphone LiDAR is good at while incorporating a system that increases absolute positional reliability.

Furthermore, if you may capture the same site multiple times, using the same references each time is indispensable. If references change between sessions it becomes hard to judge what caused differences during comparisons. For pre/post-construction comparisons, progress management, and fixed-point records, consistency of references directly improves data usability. Thinking of reducing errors in smartphone LiDAR surveying as creating data that remains meaningful when compared later—not just making a single capture look good—clarifies the importance of control points.

Tip 5: Check the point cloud immediately after acquisition and re-capture

The fifth tip is to check the point cloud and shapes immediately after acquisition and re-capture on the spot if necessary. In smartphone LiDAR surveying, even when capture seems to have gone well on site, magnifying later can often reveal distortions or missing areas. Typical issues that are hard to notice during acquisition include waviness at walls, floor inclinations, column smearing, rounding of corners, missing openings, and depth shifts. Therefore, make checking immediately after capture part of your workflow rather than stopping at measurement.

The check should not be limited to whether the whole area is captured. Inspect whether the areas relevant to the dimensions you care about are intact, whether the floor-to-wall intersections look natural, whether parts that should be straight are not curved, whether reference points are clearly captured, and whether there is excessive noise. Especially if you plan to use the data later for sections or as a base for plans, carefully check the reproducibility of corners, grid lines, and edges. Failing to check these on site can lead to major rework later for issues that could have been redone in a few minutes.

In practice, post-acquisition checks are sometimes done perfunctorily because visual presence of data on the display gives a false sense of security. What matters is checking with the intended use in mind. For example, if you are using the data for equipment installation planning, focus on corridor widths and around obstacles; if it is a base for renovation planning, check for missing areas near the ceiling and vertical rises. Having check points for each purpose makes decisions about re-capture easier.

If re-capture is necessary, it is effective to target only the problematic areas for a short re-run. Re-doing the whole area from the beginning not only takes time but can also introduce different errors. After identifying what was bad, re-acquire by changing viewpoint orientation, shortening the route, increasing control points, or re-entering from a spot with better lighting—address the cause and improvement is easier. Think of smartphone LiDAR surveying as increasing quality through iterations of acquisition and checking rather than one-off capture.

Fostering a culture of checking also prevents dependence on specific individuals. Relying only on those who can capture well makes reproduction difficult as the number of sites increases. To ensure minimum quality regardless of who measures, share the check items and set decision criteria for re-capture. The final push to reduce errors in smartphone LiDAR surveying often depends less on acquisition technique than on a strict attitude toward post-acquisition review.

Tip 6: Separate correction and validation by deliverable

The sixth tip is to separate correction and validation methods by deliverable. Data from smartphone LiDAR surveying is not equally suitable for all uses just because it was captured once. It may be sufficient for three-dimensional overviews but leave doubts for sectional comparisons; when flattened and used to draw lines, local distortions may become apparent. To reduce errors, do not lump overall data quality together—evaluate it according to intended use.

For example, if the main purpose is current-condition understanding or stakeholder sharing, some local roughness may be acceptable as long as the overall spatial relationships are visible. On the other hand, if you use the data as a base for drafting or quantity estimation, stability of reference planes, consistency of representative dimensions, and absence of oddities in sections are important. When integrating with other positioning data, alignment to a coordinate system and matching control points are prioritized over smoothness of shape. In other words, even the same smartphone LiDAR data requires different error considerations depending on the intended use.

If you do not understand these differences, you may undervalue usable data or, conversely, repurpose it where it should not be used. A common scenario on site is using data for drafting simply because it looks clean, or dismissing all data because part of it was misaligned. The important thing is to change your correction approach according to purpose. Applying global position alignment, cutting out areas to check local shape, removing unnecessary noise, or grooming the data to be section-friendly—all these tailored treatments increase the practicality of the same dataset.

Validation should be handled similarly. Decide in advance validation methods according to the deliverable: check distances between representative points, compare differences from known heights, confirm linearity in sections, or overlay pre/post-construction data using the same references. Having these checks prepared prevents unnecessary worry. Sites that use smartphone LiDAR surveying effectively do not treat all outputs as universally fit; they calmly organize what is sufficient for each use and where support is needed. That organization is the key to preventing errors from becoming practical problems.

Ultimately, what matters is maximizing value while understanding the data’s limits. Smartphone LiDAR surveying enables recording at timings previously impractical and quick situation assessments. By leveraging that strength and separating correction and validation by use, you do not have to fear errors unnecessarily nor use the data inappropriately. Improving accuracy is not only about pushing numbers; it is about bringing data into a reliable state for its intended use.

Operational rules to confirm before using on site

We have covered six tips, but what truly makes a difference in practice is whether these are institutionalized as operational rules on site rather than remaining personal experience. Smartphone LiDAR surveying is easy to start using, but that very ease makes variability between people more likely. That is why you need to establish minimum rules before deploying it on site.

The first thing to decide is which tasks to use it for and which tasks not to use it alone. Clarifying use cases—effective for short current-condition checks, interference checks, pre-renovation records, and progress sharing, but only auxiliary in situations demanding rigorous standards—helps prevent unrealistic expectations. Problems arise on site not from technical limits themselves but from ambiguous expectations.

Next, standardize acquisition procedures, even if simply. Make common rules such as: check site conditions before starting, segment measurement areas, unify walking speed and device handling, always perform post-acquisition checks, and record reference points. Standardizing does not have to sound grand; it simply means sharing likely failure points in advance so quality remains stable even if personnel change.

In the trial introduction phase, compare results with existing measurements and drawings to understand the magnitude of differences under your company’s site conditions. Errors behave differently in indoor vs. outdoor, confined vs. open spaces, and flat floors vs. areas with many steps. Rather than deciding operations based only on generalities, knowing how stable it is in the types of sites you frequently handle clarifies appropriate use and precautions. Companies that succeed with smartphone LiDAR surveying in practice carefully conduct this evaluation during initial introduction.

Finally, if you want to advance site use including positional accuracy, separate three-dimensional capture and high-precision positioning in your thinking. Smartphone LiDAR acquisition excels at shape capture, but when you need greater absolute positional reliability, coordinate referencing is essential. A practical operation is to combine the quick shape capture by smartphone with higher-precision position information. Thinking of shape capture and position fixing as separate systems makes it easier to reconcile usability and accuracy on site.

Summary

The most important point in reducing errors in smartphone LiDAR surveying is not leaving everything to the device. Decide the measurement purpose and required accuracy first, check site conditions before measurement, stabilize walking and device orientation, suppress misalignment with control points and known points, check immediately after acquisition and re-capture if needed, and separate correction and validation by deliverable. Just by following these six points, the stability of results from the same smartphone LiDAR surveying can change dramatically.

What is truly required in practice is not always obtaining the highest-precision data, but being able to consistently reproduce data with the necessary accuracy for each situation. Smartphone LiDAR surveying is highly effective for accelerating initial on-site actions, preserving records, and creating a common understanding among stakeholders. At the same time, when you need to reliably handle positional references, combining shape capture and high-precision positioning is indispensable.

If you want to make three-dimensional data captured with a smartphone more practical on site, the next step is to establish an environment that can provide high-precision positional information to the shapes captured by LiDAR. LRTK, as an iPhone-mounted GNSS high-precision positioning device, is a compatible option for sites that want to leverage smartphone mobility while increasing positional reliability. If you want to move beyond treating smartphone LiDAR surveying as merely a simple record and apply it to position checks, current-condition understanding, and construction management, consider integrating the mechanisms for shape capture and position determination. That is the most realistic path toward usable on-site three-dimensional measurement.