How to assess RTK GPS accuracy for work? Decision criteria to keep in mind by use case

When considering the introduction or operation of RTK GPS, many practitioners first worry about “what level of accuracy can be achieved.” However, what really matters in the field is not simply looking at the nominal accuracy. It is important to determine whether the positioning accuracy required by your work can be met, whether it can be operated stably with reproducibility, and whether the positioning results can be used for operational decision-making.

RTK GPS can achieve high-accuracy positioning under favorable conditions, but it is not equally stable in every field environment. While open-sky locations may yield good results, errors can increase near buildings, in heavily wooded areas, or where sky visibility is limited. Even with the same equipment, results vary depending on work procedures, communication environment, observation time, and the reflective environment around the site. In other words, assessing RTK GPS accuracy is not about reading numbers alone, but about judging use case, environment, and operation together.

This article organizes and explains the basic concepts you should understand when evaluating RTK GPS accuracy for work, the representative indicators to check in practice, considerations by use case, and points to stabilize accuracy on site. It is compiled to be practical not only for those considering introduction but also for those already using RTK GPS who are unsure how to interpret the numbers.

Table of Contents

‐ The key concepts to know first when judging RTK GPS accuracy ‐ Representative indicators to check when assessing RTK GPS accuracy ‐ How judgment criteria change by use case ‐ Why high accuracy alone is insufficient ‐ Practical verification methods to assess RTK GPS accuracy on site ‐ Main causes of RTK GPS accuracy instability ‐ Checkpoints to consider separately before introduction and during operation ‐ Summary: how to leverage RTK GPS accuracy in work

The key concepts to know first when judging RTK GPS accuracy

When evaluating RTK GPS accuracy, the first thing to understand is that “high precision” and “usable for work” are not the same. For example, even if centimeter-level positioning (several centimeters (inches)) is possible under certain conditions, it becomes hard to use operationally if that result cannot be obtained consistently. Conversely, for applications where the required operational standard is not that strict, some variation may still be perfectly useful.

What becomes important here is deciding the required accuracy first. Instead of first thinking about how high-performing the RTK GPS is, you need to clarify how many centimeters (inches) of error your operations can tolerate and judge whether that level can be continuously met. Required accuracy varies greatly depending on whether the task is current status surveying, stakeout/layout, as-built verification, or matching with an asset register.

Also, the word “accuracy” has several meanings. Sometimes it refers to getting a good position in a single observation, sometimes it emphasizes repeatability—getting nearly the same value when measuring the same point multiple times—and sometimes it refers to absolute correctness—how close the result is to the true position. Confusing these in practice leads to wrong judgments. Don’t be reassured just because a number is small; understand what that number actually indicates.

RTK GPS combines satellite signals and correction information to perform high-precision positioning, but it is affected by satellite geometry, observation environment, communication status, and reflections from surrounding structures. In other words, positioning results are highly environment-dependent. Therefore, it is risky to use the numbers given in catalogs or specifications under ideal conditions directly as a basis for operational decisions. You must assess whether the accuracy can be reproduced in your own operational conditions.

Time is also important in accuracy judgments. A good number at one moment is insufficient if results differ between morning and afternoon, cannot be reproduced on another day, or vary with different operators. What the operation needs is not a single good result but the ability to maintain a certain standard regardless of who uses it and where. Therefore, when judging RTK GPS accuracy, give more weight to stability during continuous operation than to one-off successes.

From the standpoint of practitioners, it is easier to think in terms of “Can sufficient accuracy for this use be stably ensured under site conditions?” rather than “Is this device high precision?” Adopting this perspective changes both pre-introduction evaluation methods and in-operation checks.

Representative indicators to check when assessing RTK GPS accuracy

When judging RTK GPS accuracy, you should look at several representative indicators. However, knowing only the names of these indicators is not enough. It is important to understand what each indicator means for on-site judgment.

First, be aware of the difference between horizontal and vertical components. Generally, RTK GPS tends to have different accuracies for planar position and height, with height usually disadvantaged. Some operations only require correct horizontal position, but for tasks that involve elevation, depth, or slope control, vertical stability is crucial. Do not rely solely on planar numbers; evaluate height errors separately.

Next, whether a fixed solution has been obtained is important. In RTK GPS, the stability of the solution changes depending on the state of the correction calculation. If the positioning state is not sufficiently stable, coordinates may appear but the expected accuracy may not have been achieved. In practice, people sometimes adopt coordinates without checking the solution state; this should be avoided. Knowing the state under which coordinates were obtained and keeping it in operational records makes later verification easier.

The variation when repeatedly observing the same point is also an important indicator. Even if a single observation gives a good number, you cannot call the result highly reliable if re-observation shifts by several centimeters (inches). Conversely, if you can return to the same point and get nearly the same result even after some time, it is reasonable to consider the device and operation stable. In practice, reproducibility checks by multiple observations are often simplified, but they are very effective for assessing accuracy.

Don’t forget satellite reception conditions. While more satellites are generally advantageous, judging by count alone is dangerous. What matters is whether the sky distribution is biased, whether reception from certain directions is weakened by obstacles, and whether reception varies significantly with time of day. Don’t rely only on easy-to-read numbers; judge based on the overall site environment.

Additionally, the time from initialization to stable positioning directly affects usability. High theoretical accuracy is of little use if it takes a long time to stabilize every time, as this reduces work efficiency. For tasks that require short observation times per point, initialization speed and ease of re-initialization are part of the judgment. Conversely, for tasks that observe a few points carefully, it may be better to prioritize stability even if it takes more time.

The key is not to look at any single indicator in isolation. Good horizontal accuracy with low repeatability leaves doubts, and a fixed solution in a strong-reflection environment may be misleading. Combining multiple indicators for a comprehensive judgment is essential when assessing RTK GPS accuracy. In practice, people want a single easy-to-understand number, but accuracy should be understood as the result of multiple factors.

How judgment criteria change by use case

The most important thing when assessing RTK GPS accuracy for work is to change judgment criteria by use case. Using the same accuracy standard for all operations can lead to overquality and reduced efficiency or to operating with insufficient accuracy.

For tasks such as current status surveys, locating equipment, and asset management, you can sometimes operate with relatively wide tolerances. For these tasks, it is more important to be able to identify the target on site without confusion, to understand the correspondence with maps or registers, and to be able to return to the same target on revisits than to have millimeter- or centimeter-level absolute precision. Consequently, reproducibility and practicality are weighted more heavily here than strict survey-grade accuracy. Maintaining practical positional consistency is more important than displaying high numerical precision.

On the other hand, tasks close to stakeout or layout marking require higher accuracy than simple position identification. Where placement errors affect later processes, planar errors must be viewed more strictly. For these tasks, it is important not only to get a good observation once but also to ensure that repeated use on site does not let errors accumulate. Therefore, confirming fixed solution state, multiple observations, and cross-checking with known points are part of the judgment criteria.

For as-built management and construction management tasks, vertical stability as well as planar stability is important. Where elevation management directly impacts quality—such as slopes, pavements, and around structures—height variation cannot be ignored. On site, people may feel reassured if horizontal position matches, but if vertical stability is lacking, trust in the overall control values collapses. For such uses, evaluate horizontal and vertical separately and confirm that each meets required levels.

Also, for tasks that require collecting many points in a short time, overall processing efficiency and stability are evaluation targets as well as per-point theoretical accuracy. For example, when acquiring many points over a wide area, slightly shorter observations per point can be acceptable if the overall quality meets needs. In such cases, consistency across the area and balance with work time matter more than exact single-point accuracy. Thus, productivity is part of the evaluation alongside accuracy.

Conversely, for reference points that serve as the basis for subsequent processes—such as setting control points or acquiring coordinates that later serve as the foundation—you should be cautious. Because these affect all subsequent work, a single temporarily good result is insufficient. More rigorous checks—re-observing at different times, comparing under different surrounding conditions, and verifying against known benchmarks—are required. Whether the use is for foundational data or reference information greatly affects the required level of verification.

As described, RTK GPS judgment criteria vary with the nature of the target, the impact on later processes, the difficulty of redoing work, and required work speed. It is not simply a matter of “how many centimeters is acceptable,” but rather clarifying “for which task, how much deviation is tolerable.” Practitioners should design allowable errors and verification procedures from the operational side, not just compare equipment performance.

Why high accuracy alone is insufficient

When considering introducing RTK GPS, people are often drawn to the phrase “high accuracy.” However, what truly matters in operations is not the level of accuracy itself but the ability to judge and explain the results. In other words, it is important to determine whether the positioning result can be used for on-site decisions and whether it can be justified on review.

For example, even if an excellent value appears at a point, if the reception environment, solution state, communication status, and observation time at that moment are not recorded, you cannot later verify the reliability of that coordinate. Decisions based solely on an operator’s sense of “it’s probably fine” make weak operational records. In practice, recording the rationale for adopting a result is as important as the precision itself.

Also, even highly accurate systems are useless if they are difficult to use in the field. If startup takes time, communication recovery is cumbersome, performance is extremely unstable in poor conditions, or quality varies greatly by operator, then no matter how good the theoretical performance, the system will be hard to adopt operationally. What is needed in operation is ease of producing consistent quality, keeping workflow uninterrupted, and easy judgment in abnormal situations.

Moreover, excessive precision requirements can reduce operational efficiency. Imposing long, intensive verification steps on tasks that do not require such strict accuracy increases site burden. As a result, operations may become ceremonial or verification procedures may not be followed. Accuracy management should be appropriate to the use, not simply as strict as possible.

Conversely, too lax verification for required accuracy is also problematic. For important work related to stakeout or as-built control, adopting single observations or skipping known-point checks can lead to large impacts when deviations are later discovered. Thus, what operations require is not high accuracy per se but ensuring the necessary accuracy to the necessary degree and having a system to verify it appropriately.

Here it is useful to distinguish between accuracy and quality assurance. Accuracy is the numeric result, while quality assurance concerns whether the result was obtained under reliably controlled operation. Operationally usable RTK GPS is not about occasionally getting good numbers but about understanding the conditions that produce good numbers, avoiding adoption under poor conditions, and linking verification to operational decisions. With this perspective, introduction evaluation and operational improvement proceed as operational design rather than mere equipment assessment.

Practical verification methods to assess RTK GPS accuracy on site

To assess RTK GPS accuracy in practice, catalog specs alone are insufficient. Having on-site verification procedures is important. Below are simple verification ideas that field practitioners can readily adopt.

First and foremost, verify against known positions. If there are points with known coordinates or points that can serve as comparison benchmarks, use them to check positioning tendencies. Don’t just do it once—observe multiple times over time to see how much scatter occurs; this makes judging reproducibility easier. This simple procedure is very effective for practical reliability checks.

Repeat observations of the same point are also effective. Observing the same point before and after a task and checking differences helps you grasp environmental fluctuation and operational stability that day. If there is a large difference between the start and end of the day, treat all data acquired that day with caution. Repeated observations are not only for reassurance but also for early anomaly detection.

Make it a habit to visually check sky visibility and the surrounding environment during observations. Don’t rely solely on satellite count or internal displays—check whether there are high walls nearby, many metal or glass surfaces, or the point is surrounded by trees. This helps predict risks of multipath and blockage. Near buildings or under viaducts, numbers may appear stable while actually containing biased errors.

Checking communication status is also indispensable. Because RTK GPS relies on stable reception of correction information, intermittent communication reduces positioning reliability. In the field, people tend to proceed if coordinates appear, but if communication is frequently lost and restored, handle the results cautiously. In accuracy assessment, consider communication stability as part of positioning quality.

Securing sufficient observation time also matters. In rushed sites, there is a temptation to adopt coordinates the moment they appear, but waiting a bit can allow you to see whether values stabilize. For important points, avoid making snap decisions; judge adoption based on convergence and fluctuation. Even for overall site productivity, it is effective to be more thorough for critical points.

Also, keep work records. Even simple notes on where, under what conditions, and at what time positions were acquired make it easier to review the validity of results later. This is particularly useful when abnormal values appear or when comparing to later re-measurements. Assessing RTK GPS accuracy should not end with in-the-moment judgment; it should be an operation that allows recording and rechecking.

Main causes of RTK GPS accuracy instability

When RTK GPS accuracy is unstable, the cause is not usually a single factor. In practice, people often dismiss it as “it’s just an off day,” but decomposing the causes makes countermeasures easier.

A representative factor is insufficient sky visibility. Satellite signals are easier to receive in open environments; surrounding tall buildings or trees increase blockage. In places open on only one side, reception direction bias can reduce solution stability. On site, people sometimes judge “there is a little sky visible, so it’s okay,” but it is important to assess not just the amount of sky visible but also directional bias in reception.

Next, be wary of reflections. Close-by walls, metal surfaces, vehicles, and fences can reflect signals, causing them to be received via different paths and producing errors. These errors are hard to detect visually and may appear numerically stable, making them insidious. Be especially careful when performing stakeout near buildings or when surveying around equipment.

Unstable communications also lead to degraded accuracy. If correction information reception is interrupted, the positioning state changes and result reliability decreases. Communication is not safe simply because it connected once; what matters is whether it remains stable throughout the work. In areas with weak radio signals, when working while moving, or in sites where surroundings change often, continuously monitor correction reception.

Variation in observation procedures cannot be ignored. Differences in observation time by operator, different criteria for checking solution state, or using single observations even for important points create operational variation that becomes quality differences. RTK GPS is a high-function system, but actual quality largely depends on operation; standardizing procedures is essential. Relying too much on individual experience causes personalization of skills and reduces stability.

There are also temporal variations by time of day or day-to-day. Satellite geometry, surrounding work activities, and communication congestion can change conditions even at the same location. Thus, success once does not guarantee consistent success. Cases where tests succeeded during introduction but operations became unstable in production often result from insufficient consideration of time variations.

The important point is not to blame equipment alone for unstable accuracy. By separating environment, communication, procedure, and judgment criteria as possible causes, you can see directions for improvement. Practitioners should aim not for perfect elimination of errors but for understanding conditions that cause instability and responding by tightening acceptance criteria under those conditions or using supplementary verification methods.

Checkpoints to consider separately before introduction and during operation

To properly assess RTK GPS accuracy, you need to separate pre-introduction evaluation from daily management after starting operation. Even if you carefully check before introduction, poor operation will not keep quality stable; conversely, operational effort alone cannot make an unsuitable application work.

Before introduction, clarify where in your operations you will use RTK GPS. Don’t plan to apply it uniformly to all positioning tasks; decide which tasks—current status surveys, stakeout/layout, as-built verification, asset management, etc.—are within scope. If you introduce equipment without this clarity, you may have undue expectations or restrict potential uses too much.

It is also important to test under conditions close to actual expected sites. Good results in open locations are insufficient if actual sites are mainly near buildings or trees. What matters is how stable it is at the places you most want to use it. Knowing how far you can use it under challenging conditions is more valuable for introduction decisions than success under ideal conditions.

Design verification procedures before introduction as well. Define under what conditions results are adoptable, how many times to observe important points, how often to verify against known points, and how to handle outliers. Establishing these rules makes it easier to stabilize quality after operation begins. Introducing equipment and creating operational procedures are equally important.

During operation, emphasize routine quality checks. Maintain concise but repeatable procedures such as checkpoint observations at start of day, repeated checks of important points, and end-of-day return checks. This makes it easier later to judge whether the day’s positioning was reliable.

Operator training during operation is also vital. RTK GPS is not an automatically high-precision tool; operator understanding of state confirmation and adoption judgment determines quality. Sharing how to read numbers, detect poor conditions, and identify situations where results should not be adopted reduces individual differences. Rather than a one-time briefing at introduction, periodically refine judgment criteria through operational examples.

Thus, before introduction focus on applicability and rule design; during operation focus on quality maintenance and standardizing judgments. Assessing RTK GPS accuracy for work means not only looking at performance initially but also evaluating how stably it will support operational quality over time.

Summary: how to leverage RTK GPS accuracy in work

When assessing RTK GPS accuracy for work, do not judge superiority by the size of numbers alone. What matters in practice is organizing required accuracy levels by use case and confirming that those levels can be stably met under site conditions. Separate horizontal and vertical considerations, focus on reproducibility rather than one-off good values, and account for fixed solution state, communication, and surrounding obstructions as basic elements of judgment.

RTK GPS accuracy is not determined by equipment performance only. Sky visibility, reflective environment, communication stability, observation procedures, and operator understanding all play large roles. Therefore, test under conditions close to expected sites before introduction, and continue simple verification procedures such as known-point checks and repeated observations after operation begins. Creating a condition in which RTK can be used stably in operations is more valuable than simply choosing a system that claims high accuracy.

For tasks where stakeout or as-built verification strongly affect subsequent processes, make adoption criteria for positioning results explicit. Conversely, for current status surveys and management uses, ensure that required accuracy is met while balancing usability and reproducibility. In short, assessing RTK GPS accuracy should be work-centered, not device-centered.

If you want to balance field usability and high-precision positioning, consider adopting RTK in a way that is easy to operate. For example, an iPhone-mounted GNSS high-precision positioning device like LRTK makes it easier to incorporate high-accuracy positioning into everyday field work using a familiar, easy-to-handle terminal. Rather than simply comparing RTK GPS accuracy, thinking through how it will be used, verified, and established within actual workflows increases the likelihood of realizing the benefits of introduction.