7 Causes of Large RTK Errors | Common On-Site Mistakes and Countermeasures

Even when using RTK, the expected accuracy may not be achieved. It worked fine yesterday but is off today. Points measured at different times don’t quite match when overlaid. These kinds of problems are very familiar to practitioners who use RTK on-site.

While RTK can achieve high-precision positioning, errors can suddenly appear large when conditions are not met. What makes this troublesome is that causes are often multiple rather than single. The way the sky appears, the surrounding environment, the state of correction information, installation method, coordinate settings, and operational procedures can combine so that significant offsets only appear when several factors overlap.

Therefore, to reduce RTK errors you must not only look at the device performance but also isolate causes including site conditions and work procedures. Rather than scrambling to respond after an error appears, understanding where failures are likely in advance can greatly reduce re-measurements and rework.

This article organizes seven representative causes that make RTK errors large for practitioners searching for “RTK error causes,” and explains common on-site failures and countermeasures from a practical viewpoint. If you want to understand why positioning results are unstable structurally rather than intuitively, or if you want to switch to reproducible operations, please read to the end.

Table of contents

• What to know first when RTK errors look large

• Cause 1 Poor satellite geometry and unfavorable observation conditions

• Cause 2 Insufficient sky visibility and reflections from surrounding structures

• Cause 3 Unstable reception of correction information

• Cause 4 Insufficient transition to a fixed solution and initialization checks

• Cause 5 Mistakes in antenna height and equipment setup

• Cause 6 Confusion in coordinate system or reference settings

• Cause 7 Unstandardized work procedures and weak quality control

• Practical points to reduce RTK errors on site

• Conclusion

What to know first when RTK errors look large

When considering RTK errors, the first point to note is that there are several types of errors. Simply observing “it was off by a few centimeters (a few in)” does not mean the underlying issue is the same. Offsets can consistently drift in the same direction, or positions can vary each time you measure. Sometimes only height is unstable, and other times the horizontal position is fine while elevation is incorrect.

If you apply countermeasures without understanding these differences, you may make the wrong fixes. For example, if the cause is unstable correction information but you only review the installation method, you won’t achieve a fundamental improvement. Conversely, if there is an obvious installation error but you only suspect the communications environment, you will just waste time.

Also, RTK is not omnipotent. The fact that it can provide high-precision positioning does not mean it gives the same accuracy everywhere. Only when conditions such as an open sky, few reflective objects, a stable reception of correction information, correct coordinate settings, and careful initialization checks are met will it tend to exhibit its true performance.

In practice, when errors occur they are sometimes dismissed with “the satellites were acting up today.” However, in many cases the cause lies more in site selection and operational procedures than the satellites themselves. In other words, the key to reducing errors is not only a deep technical understanding of positioning mechanisms but also anticipating and eliminating points where failures are likely in the field.

Below we look at seven causes that can make RTK errors large. Sometimes only one applies, but often multiple factors overlap. As you read, consider which patterns are likely at your sites so you can more easily identify improvement points.

Cause 1 Poor satellite geometry and unfavorable observation conditions

One indispensable cause of large RTK errors is satellite geometry. It is often assumed that more satellites automatically mean good results, but what matters is not only the number but where the satellites are distributed in the sky. If the distribution is biased, positioning stability can decline even if satellites appear observable.

For example, if observed satellites are concentrated in a particular direction, geometric conditions worsen and the solution becomes weak. As a result, errors tend to occur in horizontal and vertical directions. Height in particular is susceptible; even if horizontal positioning appears fine, elevation can look unstable.

A common on-site mistake is checking only that a value appears immediately after starting positioning and then proceeding with work. Even if a value is displayed, if satellite geometry is unfavorable at that time, the result may be less stable than it appears. Sites that try to save time are especially prone to skipping this check, but rushing here can lead to major rework downstream.

As a countermeasure, be mindful of observation conditions before starting work. In addition to ensuring the sky is sufficiently open, calmly confirm the state just after starting positioning and avoid hastily recording the first point. If you work at the same site over multiple days, it’s useful to identify time windows with relatively good conditions. For tasks with strict accuracy requirements, fixing the time of day so observations occur under similar conditions each time can improve reproducibility.

Another important perspective is recognizing anomalies. If a site was fine until yesterday but became unstable today, before concluding the instrument is faulty you should suspect differences in observation conditions. Signs such as not reaching a fixed solution quickly, an unstable solution, or large scatter in repeated observations of the same point indicate that observation conditions, including satellite geometry, may have deteriorated.

RTK is built on precise repeated observations. Knowing that even visually similar blue skies can be good or bad for positioning will markedly change on-site decisions.

Cause 2 Insufficient sky visibility and reflections from surrounding structures

A representative on-site factor that increases RTK errors is insufficient sky visibility and signal reflections. Near mountains, below slopes, next to buildings, in areas with many trees, around metal equipment, or where there are many vehicles, satellite signals may be blocked or reflected instead of arriving directly. Observing in these conditions often causes unstable positions, difficulty achieving fixes, and poor reproducibility when measuring the same point.

Reflections are particularly important to watch for. When signals arrive via bouncing off walls, metal surfaces, glass, or machinery instead of direct paths, the received signal has taken a longer route than intended, which results in positioning errors. Since reflections are invisible, the person measuring may mistakenly assume “the sky is a bit visible so it’s okay,” which is troublesome.

A common failure on site is judging based only on what is directly above the point to be measured. Even if the sky right above is slightly open, observation conditions are not adequate if there are many tall obstructions around. Especially at construction or infrastructure sites, temporary reflectors such as temporary materials, heavy equipment, stored materials, fences, and scaffolding can exist only for that day. If it was fine yesterday but is off today, suspect changes in such surrounding conditions.

As a countermeasure, revisit how you choose measurement positions. Even when measuring the same target, changing your stance slightly can improve conditions. In places where obstructions are unavoidable, rather than lingering at that exact spot, plan work by organizing relationships with nearby control points that have better conditions; this usually produces more stable accuracy.

When vertical errors are large, adopting the habit of suspecting reflections or blockages is useful. On-site attention tends to focus on horizontal offsets, but vertical discrepancies can be more serious in practice. In construction management and as-built verification, a height difference of a few centimeters (a few in) can change decisions. That is why you should not underestimate the influence of sky visibility and reflections.

Also, verbalizing and confirming the surrounding environment before starting observations is effective. If you share conditions like many trees, tall buildings on one side, nearby metal surfaces, or frequent heavy equipment movement with your team, causes will be easier to trace when anomalies occur. RTK errors are not only caused by the machine’s internals but also by the environment the site creates.

Cause 3 Unstable reception of correction information

RTK uses correction information to achieve higher accuracy than standalone positioning. If this correction information is not stably received, even excellent onboard observation conditions will not reach the expected accuracy. A surprisingly common on-site mistake is focusing only on satellite status and postponing checking the state of correction information.

There are several causes for unstable correction reception. Examples include being in a place with poor communications, radio congestion, connection state changing while moving, and the influence of surrounding terrain or structures. These problems can interrupt or delay correction updates and affect solution stability.

A common on-site mistake is proceeding simply because a display that looks like a fixed solution appeared. Even if correction reception is intermittent, it may briefly appear acceptable. But if you increase measured points in that state, positions may gradually drift in later overlays, or points measured at different times may not align with point clouds or drawings. This often goes unnoticed during fieldwork and is only discovered after returning to the office—this is a typical failure.

As a countermeasure, treat communication quality as part of site conditions. Not only sky visibility but whether correction information can be stably received is directly linked to observation quality. If communications are unstable, plan re-observation rules and increase verification points from the start. For example, for important points, measure repeatedly at different times, frequently check alignment with reference points, or return to known points at set intervals.

It is also important to create a site culture that does not downplay the state of correction information. When RTK errors occur, discussions often focus on device sensitivity or satellite numbers, but many cases are simply due to unstable correction reception. In locations with questionable communications, making a risk-aware work plan from the outset reduces unexpected rework.

RTK is not a system that completes by only receiving signals from satellites. High precision is realized only when the correction information that supports it is stably received. Making checking correction information a habit is fundamental to accuracy management.

Cause 4 Insufficient transition to a fixed solution and initialization checks

A very common RTK operational error is starting measurements with lax checks on solution status. In practice, time pressure leads people to assume “I can measure once some position appears on the screen.” However, in RTK it is crucial to confirm whether the solution has stably transitioned to a fixed solution and whether that state continues, and skipping this step leads to large errors.

If fixed-solution confirmation is insufficient, coordinates that superficially look plausible may still be internally unstable. As a result, the first few points may be oddly off, or the start of a line and its later part may not align. This mistake is likely when observing first thing in the morning, after power cycling, or immediately after moving a long distance.

A common site mistake is making an important point the first point right after initialization. You should wait until the state stabilizes, and even when you should recheck the same point or verify consistency with surrounding points, people often rush into work. If that first point becomes the basis for subsequent work, the error persists to the end. When the reference is off, everything that follows appears off, so the damage can be large.

As a countermeasure, integrate solution status checks into the work flow. Rather than relying on the operator’s experience or intuition, formalize what to check at the start of observations and how stable the solution must be before measuring important points. This reduces variability in decisions. On sites with multiple people, different individuals’ standards for “it’s OK now” can vary and destabilize quality.

Using known points and recheck points is also effective. Instead of going straight to unknown points after initialization, first check at an easy-to-verify point to confirm the state before proceeding to the main work. Skipping checks to save time often leads to re-measurements and overall increased time.

RTK may seem to produce high precision instantly when observing only results. But achieving stable accuracy in practice requires carefulness in the first few minutes. Not being sloppy about fixed-solution confirmation and post-initialization judgment is a basic action to prevent errors.

Cause 5 Mistakes in antenna height and equipment setup

While satellite and communication issues attract attention regarding RTK errors, human setup mistakes are also a very common cause. Typical examples include antenna height input errors, non-vertical poles, unstable mounting, and incorrect positioning over observation marks. These may seem simple but occur frequently on sites.

Antenna height input errors often appear as vertical errors. Causes include typos, unit misunderstandings, different measurement methods, and forgetting to change settings. Even if the operator thinks they entered values correctly, confirmations may be skipped in a busy environment. Moreover, vertical errors are harder to notice than horizontal ones and are sometimes only found later during cross-sections or as-built checks.

Pole tilt must not be overlooked. Even when the tip is aligned with the mark, if the pole is tilted the antenna position higher up is displaced. Small tilts can accumulate into non-negligible offsets. Poles are harder to set stably on soft ground, slopes, narrow spots, or windy days, and accuracy can vary even for the same operator.

Ambiguous mounting over the observation mark is another issue. When targeting pavement corners, boundary markers, nails, or temporary points, slight differences in what each operator considers the center of the point will result in offsets independent of RTK. Even with high-precision devices, rough point alignment produces poor results.

As a countermeasure, turn installation into procedures rather than skills. Fix a flow of measuring antenna height, entering it, verbally confirming it, and recording it so work is not ambiguous. Pole verticality checks are often skipped by experienced operators who rely on feel, but consistently confirming by the same standard each time is important.

For important points, repeatedly observe to visualize setup-derived variability. If results fluctuate over a short time at the same point, suspect not only satellites or corrections but also setup reproducibility. Although installation mistakes are often blamed on equipment, they are also an area where improvements yield large effects. Careful setup is unglamorous but the quickest route to raising RTK accuracy.

Cause 6 Confusion in coordinate system or reference settings

Some causes of apparent “large errors” on site are not positioning errors at all but differences in coordinate settings. In other words, RTK itself may be functioning normally, but if the comparison coordinate system or reference differs, results appear shifted. This is common and can be troublesome to detect.

Typical examples include using the wrong zone number of plane rectangular coordinate systems, inconsistencies with the coordinate reference of existing drawings, differing vertical datum interpretations, and confusion with local site coordinates. If the horizontal appears plausibly correct but the entire set is shifted in one direction, or if elevation differences are consistently present, or data from another team does not align, you should first suspect setting inconsistencies.

A frequent on-site mistake is starting work with the previous settings still loaded. Carrying over settings from another site or starting without thoroughly checking the drawing’s coordinate reference leads to continuing the first few points without noticing problems. Since discrepancies often appear only when overlaid with other data later, tracing the cause takes time.

When multiple people work together, not everyone necessarily understands the same reference. One person may speak in site-local coordinates while another uses a public coordinate system. These mismatches are unrelated to RTK performance but can appear as significant errors.

As a countermeasure, always organize references before work. Record in writing what coordinate system will be used, what vertical datum applies, which standard the deliverables should follow, and how to align with existing drawings or data. Avoid leaving coordinate checks to individual operators; instead include them in a checklist.

Also, compare measured results with known points or existing data in the early stages to see if there are systematic offsets. If a systematic difference appears, it is more likely a setting error than a satellite or communication issue. Distinguishing between observation accuracy problems and setting consistency problems speeds up troubleshooting dramatically.

While RTK error countermeasures tend to focus on site environment and equipment operation, managing coordinate systems and references is equally important. High-precision positioning results are useless if the reference is mistaken.

Cause 7 Unstandardized work procedures and weak quality control

The final cause of large RTK errors is not a single technical factor but the overall operation: unstandardized work procedures and lax quality control. In the field, experienced operators may seem to run things well in their own way, but such an approach lacks reproducibility and accuracy tends to vary when personnel change.

For example, one operator always takes a verification point before starting observations while another omits it. One waits a set time after initialization while another proceeds immediately. If methods for checking antenna height, re-observation criteria, stop criteria for anomalies, and how results are recorded are not standardized, results will not be stable even with the same equipment.

A common failure is postponing standardization until a problem occurs and only then investigating causes. That approach allows the same troubles to recur at different sites. Because RTK equipment often produces seemingly acceptable values even with sloppy operation, problems are not obvious, and by the time they are noticed a wide range of data may already have been collected.

As a countermeasure, do not rely on individual skill for quality control. Standardize check items at observation start, rules for re-observing important points, frequency of known-point verification, how to handle outliers, and how to keep records so that anyone will make similar judgments. This reduces not only variability in accuracy but also time spent on cause investigation.

How you review incidents when errors occur is also important. Don’t just conclude “accuracy was poor today”; instead, each time organize whether the issue was satellite conditions, sky visibility, correction reception, setup, settings, or procedures. Doing so directly leads to improvements next time. These small on-site learnings accumulate into large differences in overall quality.

RTK accuracy is not decided solely by machine performance. No matter how precise the system, unstable operations produce unstable results. To reduce sites with large errors, work on operational flow as well as technical understanding.

Practical points to reduce RTK errors on site

We’ve covered seven causes, but what is really important in practice is shifting from ad hoc responses each time an error appears to operating initially in a way that prevents errors. For that, keep some practical on-site points in mind.

First, evaluate site conditions before using RTK. Rather than discovering instability after starting, checking sky visibility, reflectors, communications, footing conditions, and whether known points are available helps you avoid infeasible measurement approaches. When entering a site for the first time, look beyond the target and consider the surrounding environment.

Next, treat the first point carefully. Don’t take the initial positioning result right after starting or after moving as the baseline; check stability with verification points, allow time as needed, and repeat the same point. These basic actions greatly affect overall quality. Rushing the first few minutes can cost many hours later.

Also, double- and triple-check important points. Don’t finish with a single pass for boundaries, as-built management points, or points that will be used as references later. Re-measure at different times, confirm from different directions, and check relationships with known points to reduce oversights.

Keep site records. Record what time points were measured, surrounding conditions, correction states, which settings were used, and where anomalies occurred. Reproducible operations come from records, not memory.

If working as a team, create standard procedures. Sites run by individual experience may hide problems while things are going well, but accuracy differences surface when personnel change. Align check procedures so anyone can achieve minimum quality; this is one of the most effective practical countermeasures.

Finally, adopt the mindset that RTK errors are not eliminated but managed. It is difficult to create ideal conditions in every site, but knowing what is risky, where to check, and when to stop on encountering anomalies prevents major rework. Rather than fearing errors, understanding and controlling how they manifest leads to practical RTK operation.

Conclusion

There is not a single cause for large RTK errors. Satellite geometry, sky visibility, reflections, correction reception, insufficient fixed-solution checks, antenna height and setup mistakes, coordinate system confusion, and variability in work procedures all combine to produce errors. Therefore, when an error occurs you should not suspect only the device but isolate causes including site conditions and the entire operation.

To achieve reproducible accuracy in practice, beyond using good equipment you must know in advance the situations that tend to produce errors and not omit verification procedures. Careful judgment immediately after starting positioning, choosing observation positions, rechecking important points, ensuring setting consistency, and thorough work records will greatly improve RTK stability.

If you want to use RTK more reliably on-site, in addition to high-precision equipment choose a system that is easy for anyone to use, easy to verify, and operable reproducibly. With the iPhone-mounted GNSS high-precision positioning device LRTK, on-site positioning work becomes more accessible while facilitating the use of high-precision location data. If RTK errors trouble you and you want simpler, more reliable on-site handling, consider examining how to proceed with simplified surveying using LRTK.