top of page

RTK Accuracy Explained: ±2 cm in Real Life? Factors, Fix vs Float & Quality Checks

By LRTK Team (Lefixea Inc.)

All-in-One Surveying Device: LRTK Phone

Table of Contents

What is RTK?

Accuracy Achieved with RTK Positioning

Factors Affecting RTK Accuracy

Key Technologies Supporting RTK Accuracy

Use Cases of High-Precision RTK

Simple Surveying with LRTK

FAQ


What is RTK?

Global Navigation Satellite Systems (GNSS) are now used in a wide range of fields, from logistics management to smartphone mapping apps. However, standalone positioning with ordinary GPS/GNSS typically has errors of about 5-10 m (16.4-32.8 ft). This is insufficient for applications that require high positional accuracy, such as autonomous vehicles or precision construction surveying. One technology that addresses this need is Real-Time Kinematic (RTK) positioning. RTK receives GNSS signals simultaneously at two points—a reference station (base) and a rover—and applies error correction information from the reference station to the rover, reducing positioning errors to the order of a few centimeters (a few inches).


The basic principle of RTK is differential positioning. A reference station is a receiver whose precise coordinates are known in advance and that continuously receives GNSS satellite signals. A rover (such as a work vehicle or a handheld receiver) receives satellite signals while moving. By comparing data received from the same satellites at both locations, common error sources such as ionospheric and tropospheric delays and satellite orbit errors can be canceled out. For high-precision GNSS positioning, measurements using the carrier-phase of the satellite signal are particularly important. By resolving the minute wavelength-level differences of the carrier (integer ambiguity resolution), it is possible to compute positions with very fine accuracy on the order of centimeters. RTK refers to the method of obtaining fast, high-precision real-time positioning solutions by canceling errors through relative positioning with a reference station.


Accuracy Achieved with RTK Positioning

When RTK positioning is used under favorable conditions, positioning errors can be kept within a few centimeters (a few inches) in both horizontal and vertical directions. Typically, horizontal accuracy of about 1-2 cm (0.4-0.8 in) and vertical accuracy of about 2-3 cm (0.8-1.2 in) can be achieved. This is orders of magnitude more accurate than standalone positioning errors (several meters or more). For example, regular GPS can be off by 3-10 m (9.8-32.8 ft), but RTK can reduce that to roughly one-hundredth of that error. Centimeter-level accuracy (cm level accuracy (half-inch accuracy)) is indispensable for tasks such as as-built verification in civil engineering and precision surveying.


However, RTK is not perfect in all situations. RTK accuracy is affected by various conditions. When a sufficient number of satellites are visible, high accuracy is obtained stably, but accuracy degrades when satellite geometry is poor. In urban canyons surrounded by high-rise buildings or in forested mountain areas, satellite signals may be partially blocked or reflected by buildings and terrain (multipath), and RTK can temporarily produce positioning errors reaching tens of centimeters (tens of inches). Furthermore, if the RTK-critical carrier-phase integer ambiguity resolution (the fixed solution) is unstable, the solution becomes a so-called “float solution,” reducing accuracy to a range from under 10 cm (under 3.9 in) up to several tens of centimeters (several inches). To maximize the benefits of RTK, a suitable combination of environment and technology is required.


RTK accuracy specifications also vary depending on receiver performance and the distance to the reference station. A typical high-performance GNSS receiver may specify, for example, “horizontal accuracy: 8 mm (0.31 in) ± 1 ppm; vertical accuracy: 15 mm (0.59 in) ± 1 ppm.” Here, “ppm” is an error term proportional to baseline length, meaning that accuracy slightly degrades as the distance between the reference station and the rover increases. In practice, RTK performs best within roughly 10–20 km of the reference station, and errors tend to increase gradually beyond that range. In Japan, public reference station networks such as the Geospatial Information Authority’s continuous GNSS network are deployed nationwide, and correction information from nearby stations is often available. Therefore, in typical surveying work the distance to the reference station rarely becomes a critical issue, and RTK can practically be expected to provide centimeter-level accuracy (cm level accuracy (half-inch accuracy)) in most cases.


Factors Affecting RTK Accuracy

RTK positioning accuracy is influenced by multiple factors including satellite environment and equipment settings. Below are the main elements that affect RTK accuracy.


Satellite count and geometry (DOP): It is important that the rover receiver can see a sufficient number of satellites. Generally, about five satellites are required to initialize RTK, and 7–8 or more simultaneous satellites are desirable for stable accuracy. If satellites are unevenly distributed (for example, clustered in one part of the sky), the Dilution of Precision (DOP) worsens and positioning errors increase. A multi-GNSS receiver that supports not only GPS but also GLONASS, Galileo, QZSS (Michibiki), BeiDou, etc., can increase the number of visible satellites and improve geometry, which is advantageous for maintaining accuracy.

Atmospheric effects such as the ionosphere and troposphere: The state of the ionosphere and troposphere that GNSS signals pass through before reaching the ground affects accuracy. During periods of high solar activity, ionospheric disturbances can introduce variability in signal propagation, resulting in residual errors that corrections may not fully eliminate. Tropospheric delays caused by temperature, pressure, and humidity changes also affect positioning calculations. RTK corrects these errors using models and observational data, but the farther the rover is from the reference station, the greater the atmospheric differences and the larger the uncorrected residuals may become. Thus, long baselines (tens of km or more) can slightly degrade RTK accuracy due to ionospheric and tropospheric effects.

Multipath (reflections): When GNSS signals reflect off building facades, the ground, or water surfaces, reflected waves arriving later interfere with direct waves and cause positioning errors. This multipath effect is a major source of RTK degradation in urban and mountainous areas. High-performance GNSS antennas (such as choke-ring antennas or antennas with built-in multipath filters) are designed to reduce the impact of reflected waves. Receivers also implement signal processing to reject suspect reflected data and algorithms that monitor per-satellite signal strength to detect poor measurements. In multipath-prone environments, placing the antenna in an open location and avoiding nearby reflective objects are basic mitigation measures.

Receiver performance and settings: The performance of GNSS receivers and antennas directly affects accuracy. Dual-frequency receivers can better correct ionospheric delays and obtain fixed (integer) solutions faster and more stably than single-frequency receivers. Modern high-performance receivers feature powerful processing and firmware robust against cycle slips (phase jumps caused by signal interruptions), making solutions less likely to be lost during motion. Receiver settings, such as RTK filter parameters and initialization conditions, also impact quality. For example, applying stricter criteria for declaring a fixed solution reduces the risk of false fixes but can increase initialization time or make acquiring a fix more difficult in unstable environments. Tuning settings for stability versus speed according to the application is an important aspect of RTK operation.

Communication environment: RTK requires real-time receipt of correction data from the reference station, so communication link stability is important. Network RTK over the Internet commonly uses cellular networks, but in mountainous or underground areas radio signals may be unavailable and corrections may be temporarily interrupted. If communication is lost and corrections are not received, RTK solutions can fail to maintain high accuracy within seconds to tens of seconds (fixed solutions degrade to float). To mitigate this, receivers may automate reconnection to correction streams or include a “hold” function that maintains accuracy for short interruptions using previous correction data. In Japan, receivers capable of receiving QZSS (Michibiki) CLAS correction signals via satellite communication can continue to receive corrections even outside cellular coverage. Ensuring redundancy in correction communications and satellite reception helps maintain stable RTK accuracy.


As described above, RTK positioning accuracy is determined not only by technical specifications but also by environmental conditions and system-wide design. To obtain reliable RTK accuracy, it is important to address these factors and choose optimal equipment and methods.


Key Technologies Supporting RTK Accuracy

Various technical measures are implemented to realize centimeter-level RTK positioning and sustain its quality. The main points are as follows.


Multi-GNSS and multi-frequency support: As noted above, using multiple satellite systems helps ensure a sufficient number of satellites and improves accuracy and reliability. Current high-precision GNSS receivers typically support not only GPS but also Russia’s GLONASS, Europe’s Galileo, China’s BeiDou, and Japan’s QZSS (Michibiki). This makes it easier to avoid situations where no satellites are visible in urban canyons or mountains. Supporting multiple frequencies (such as L1/L2 or L5) enables ionospheric delay correction, shortening initialization times and speeding re-fix. For example, single-frequency receivers may take minutes to achieve an initial fixed solution due to ionospheric errors, whereas dual-frequency receivers can significantly reduce that time and be more resilient to environmental changes.

Network RTK and correction services: Services that provide correction data from existing reference station networks remove the need to install your own base station. By using network RTK services distributed over the Internet via Ntrip protocols (such as VRS), users can receive real-time corrections from virtual reference points near them anywhere in the country. This minimizes the distance to a reference and allows centimeter-level accuracy to be maintained stably over wide-area operations. Additionally, some regions offer SBAS (Satellite-Based Augmentation Systems) or PPP (Precise Point Positioning) services that provide precise orbit and clock corrections. Japan’s CLAS is effectively a PPP-RTK system that delivers nationwide correction information via satellite while also accounting for regional errors, enabling RTK-like high accuracy with standalone reception. The development of such correction services has made RTK positioning easier to use without dedicated infrastructure.

Tilt compensation and INS integration: If the GNSS antenna is tilted, the measured ground coordinates will be offset. Traditionally, survey poles had to be kept vertical during measurements. Recent high-precision GNSS receivers include IMUs (inertial measurement units) that automatically compensate for tilt. By detecting the antenna’s tilt angle in real time and correcting the measured position vertically, accurate ground positions can be obtained even if the pole is somewhat tilted. Tilt compensation makes it easier to measure near building walls or on slopes, greatly improving work efficiency. IMUs also contribute to position maintenance through GNSS/INS integration; when satellites are temporarily lost—such as entering a tunnel—the IMU can provide dead-reckoning to bridge short gaps and enable a smooth return to a high-precision solution when satellites are reacquired. These inertial sensor integration technologies play an important role in making RTK positioning less interruption-prone.

High-performance antennas and quality control: Maximizing RTK accuracy requires high-quality GNSS antennas and peripheral equipment. Reference stations should use high-grade antennas with stable phase centers, and rover antennas should be sensitive and resistant to multipath. Installation location matters too; place equipment in open, unobstructed areas and secure it against vibration and tilting. Monitoring positioning data quality is also essential for reliability. Operational quality control includes monitoring RTK solution internal indicators (such as residual magnitudes and fixed-solution ratios) and discarding or remeasuring points when anomalies appear. By ensuring quality through both technical means and operational practices, reliable RTK accuracy is achieved in the field.


Use Cases of High-Precision RTK

High-precision positioning made possible by RTK is applied across many fields. Representative use cases include:


Machine guidance and automated construction: RTK-GNSS systems are increasingly installed on construction machinery such as bulldozers and excavators to control working positions in real time. With centimeter-level positioning (cm level accuracy (half-inch accuracy)), operators can precisely know the position of the bucket tip and perform excavation and grading according to design plans. This enables high-precision construction without relying solely on operator skill, shortening project schedules and improving quality. Fully automated construction machinery is also being considered for the future, and RTK’s accuracy and reliability are essential for that realization.

Drone surveying and aerial photogrammetry: Drones (UAVs) equipped with RTK can geotag aerial images with high-precision position information at the time of capture. Traditionally, many ground control points (GCPs) had to be placed on the ground to achieve high accuracy in photogrammetric mapping. RTK-equipped drones greatly reduce the need for GCPs and can yield survey products with several-centimeter accuracy (several inches) from aerial images alone. This shortens work time across large survey sites and enables safe measurement of hazardous areas where people cannot enter.

Precision agriculture: RTK’s high precision is also effective in agriculture. By installing RTK-GNSS on tractors and rice transplanters and using automatic steering, “precision agriculture” improves efficiency and precision. Accurate control of row spacing and seeding positions reduces wasted land and can improve yields. RTK can also be used to detect subtle field gradients for automated irrigation and drainage optimization. Because RTK allows travel with errors of a few centimeters (a few inches) over large fields, it is seen as a technology that helps address labor shortages.

Infrastructure inspection and public surveying: RTK’s high precision is useful for maintaining infrastructure such as roads, bridges, and railways. It is used to accurately record crack locations during deterioration inspections, measure post-disaster ground subsidence, and verify as-built conditions for roadworks, where centimeter-level accuracy (cm level accuracy (half-inch accuracy)) is often required. High-precision measurements that once took time with optical total stations can increasingly be done efficiently with RTK-GNSS. The ability to obtain results in real time is especially valuable for disaster response and emergency inspections that require immediate on-site decisions.


As shown above, reliable RTK accuracy is not merely theoretical but a practical driving force that solves diverse on-site challenges. Advances in technologies that support high-precision positioning have made previously difficult tasks possible, dramatically improving safety and efficiency.


Simple Surveying with LRTK

While advanced RTK positioning is attractive, operating it in the field has traditionally required expertise and expensive surveying equipment, which can be a high barrier. LRTK is a solution developed to make RTK’s precise positioning easier to use. LRTK realizes “simple surveying” by combining a smartphone with a small high-precision GNSS receiver.


Specifically, by using a small GNSS receiver that connects to a smartphone and a dedicated app, anyone can perform centimeter-level positioning easily. For example, with LRTK a monopod-mounted smartphone receiver can perform single-operator surveying that previously required two people. The dedicated app automatically corrects for antenna height (pole height) and averages measurements, enabling high-precision data acquisition without specialized operation. While regular GPS positioning had errors of about 5-10 m (16.4-32.8 ft), LRTK can determine positions with horizontal and vertical accuracies of a few centimeters (a few inches), comparable to professional surveying equipment. Complex setup and bulky equipment are unnecessary—just a smartphone and a pole let you obtain coordinates of desired points on site immediately.


LRTK contains a high-performance multi-GNSS, dual-frequency receiving engine and maintains high accuracy by receiving real-time corrections from network RTK services or QZSS (Michibiki) CLAS signals. Tilt compensation functions described earlier can also be used, allowing automatic correction of measured points even when the pole is not perfectly vertical, simplifying surveying in confined spaces. By integrating these modern technologies, LRTK hides the complexity of RTK from the user, making high-precision GNSS surveying easy to introduce in-house without veteran surveyors, thus reducing costs and improving operational efficiency.


As the reliability of positioning data becomes increasingly important, easy-to-use and reliable surveying solutions like LRTK are expected to be adopted in many field applications. Embedding RTK-supporting technologies into familiar tools will further expand the use cases of positional information.


FAQ

Q1. What is the difference between RTK and regular GPS positioning? A1. Regular GPS (GNSS) positioning computes position from satellite signals received by a standalone receiver, while RTK observes satellite signals simultaneously at both a reference station and a rover and uses correction information from the reference station to cancel errors. This allows RTK to improve typical GPS errors of about 5–10 m (16.4–32.8 ft) down to the order of a few centimeters (a few inches).


Q2. What equipment and environment are needed to start using RTK? A2. To use RTK you need a high-precision GNSS receiver (rover) and correction data from a nearby reference station. The reference station can be self-installed or you can obtain data from public networks such as the Geospatial Information Authority’s stations or from commercial correction services. You also need a communication method for the rover to receive corrections (radio modem or Internet connection). Recently, compact systems that pair with smartphones (e.g., LRTK systems) have appeared, making high-precision positioning possible without dedicated equipment.


Q3. Can RTK really achieve centimeter accuracy? A3. Under appropriate conditions, RTK generally achieves horizontal accuracy of about 1–2 cm (0.4–0.8 in) and vertical accuracy of about 2–3 cm (0.8–1.2 in). However, these figures represent ideal cases with good satellite visibility and minimal interference. In places like urban canyons or dense forests where signals are unstable, RTK can temporarily produce errors of tens of centimeters (tens of inches). Even so, RTK remains far more accurate than standalone positioning.


Q4. Why does RTK accuracy degrade as you move away from the reference station? A4. As the distance from the reference station increases, errors that can be treated as common between the two locations—such as ionospheric and tropospheric delays—become less similar. While these errors are largely common at short distances, differences in atmospheric conditions over tens of kilometers can introduce residual errors that corrections cannot fully remove. Consequently, the longer the baseline, the less effective RTK corrections become and accuracy slightly degrades. Network RTK mitigates this by interpolating corrections from multiple surrounding reference stations.


Q5. What happens to RTK positioning inside tunnels or outside communication coverage? A5. Inside tunnels, satellite signals cannot be received, so RTK positioning must be suspended temporarily. However, modern receivers may include hold functions that maintain a pre-entry fixed solution for several seconds and bridge functions using inertial navigation to estimate short-term movement. These features can maintain high accuracy across short tunnels. Regarding loss of communication, in Japan receivers capable of receiving QZSS (Michibiki) CLAS signals can continue to receive corrections without Internet access. If correction data is completely unavailable for a prolonged period, maintaining high accuracy becomes difficult, but short interruptions can often be handled with these measures.


Q6. Can non-experts use RTK positioning? A6. Traditional RTK systems required specialist knowledge and equipment operation, but user-friendly solutions are increasingly available. Systems like LRTK, which use smartphone apps to automate corrections and measurements, allow people without surveying experience to obtain centimeter-level positioning easily. While basic GNSS knowledge is beneficial, it is now much easier than before to implement high-precision positioning. Even beginners can achieve reliable RTK accuracy by following manuals and standard procedures.


Next Steps:
Explore LRTK Products & Workflows

LRTK helps professionals capture absolute coordinates, create georeferenced point clouds, and streamline surveying and construction workflows. Explore the products below, or contact us for a demo, pricing, or implementation support.

LRTK supercharges field accuracy and efficiency

The LRTK series delivers high-precision GNSS positioning for construction, civil engineering, and surveying, enabling significant reductions in work time and major gains in productivity. It makes it easy to handle everything from design surveys and point-cloud scanning to AR, 3D construction, as-built management, and infrastructure inspection.

bottom of page