Latest Trends in RTK Accuracy: Evolving High-Precision Positioning Technologies

Table of Contents

• What RTK accuracy is

• Accuracy improvement through multi-GNSS and multiple frequencies

• Proliferation of network RTK and the emergence of PPP-RTK

• Miniaturization of high-precision GNSS equipment and easier use

• Further accuracy enhancement through sensor fusion and AI

• New application fields for RTK technology

• Simple surveying with LRTK

• FAQ

What RTK accuracy is

RTK (Real-Time Kinematic) is a technology that achieves high-precision GNSS positioning in real time. Normally, when receiving signals from GNSS satellites such as GPS alone, positional accuracy is limited to on the order of several meters. However, by using RTK positioning and correcting error sources, it is possible to improve accuracy to the centimeter level. Specifically, the signals received simultaneously by a reference station and a rover are compared, and errors that occur as radio waves pass through the atmosphere or due to satellite orbit errors are corrected in real time. Through this mechanism, very high accuracy positioning of approximately 2-3 cm horizontally and approximately 4-5 cm vertically is generally achieved (horizontal approximately 2-3 cm (0.8-1.2 in), vertical approximately 4-5 cm (1.6-2.0 in)).

High-precision positioning is increasingly important across a wide range of fields such as surveying, civil engineering and construction, agriculture, and autonomous driving. For example, in as-built control at construction sites or infrastructure inspections, deviations of a few centimeters can greatly affect quality and safety. Therefore, technologies like RTK that can determine positions to the centimeter are indispensable. This article explains the latest technologies and trends that underpin such RTK accuracy, and introduces how high-precision positioning technologies are evolving.

Accuracy improvement through multi-GNSS and multiple frequencies

One major factor improving RTK accuracy is the increase in available satellites. Traditionally, mainly US GPS satellites were used, but in recent years the use of multiple satellite constellations (GNSS) such as Russia’s GLONASS, Europe’s Galileo, China’s BeiDou, and Japan’s QZSS (Michibiki) has progressed, enabling “multi-GNSS.” With more satellite options in the sky, it has become easier to secure a sufficient number of satellite signals even in urban areas with poor visibility or mountainous regions, greatly improving the stability and reliability of positioning. An increase in satellite count also improves positioning geometry, enabling high-precision RTK positioning in environments that were previously difficult.

Furthermore, support for multiple frequencies (multi-band) is also increasing RTK accuracy. GNSS satellites transmit signals on multiple frequencies such as L1 and L2. Modern high-precision GNSS receivers can receive and use these multiple frequencies simultaneously to cancel frequency-dependent errors like ionospheric delay and to enable faster integer ambiguity resolution (resolving carrier-phase cycle ambiguities). For example, error factors that took time to resolve with L1 alone can be corrected quickly by dual-frequency L1/L2 reception. As a result, the time to obtain an initial fixed solution (integer solution) is shortened and the availability of real-time positioning improves. In addition, next-generation satellite signals (such as GPS L5 and Galileo E5) have wider bandwidths and designs that are more resistant to multipath (errors caused by reflections), and support for these new frequencies also contributes to accuracy improvements.

In summary, the increase in visible satellites due to multi-GNSS and the improved error correction capability from multi-frequency support have dramatically enhanced RTK positioning accuracy and stability. This is particularly important for overcoming challenges such as “not being able to acquire satellites in city canyons” or “unstable accuracy due to ionospheric effects.”

Proliferation of network RTK and the emergence of PPP-RTK

To achieve RTK accuracy, not only satellite signals but also correction information are indispensable. Traditionally, users would set up a base station at a known coordinate and exchange correction data with the rover via radio communication—the “local base station method.” Recently, however, the use of network-based RTK (Ntrip method) has spread rapidly, eliminating this hassle. By using services that distribute correction information over the Internet from CORS networks (continuously operating reference stations) maintained by national or local governments, users can perform RTK positioning without preparing their own base stations. For example, in Japan, VRS-type correction services based on the Geospatial Information Authority of Japan’s CORS network are provided, allowing rovers to receive real-time error corrections via mobile communications. Network RTK has largely eliminated accuracy degradation due to baseline length (distance to a base station), enabling uniformly high-precision positioning across the country. However, this method still requires a contract with a service provider and a communications line, so usage costs and inability to operate outside coverage remain constraints.

Amid this, a new correction technology attracting attention is PPP-RTK. PPP-RTK (Precise Point Positioning - Real Time Kinematic) distributes error information computed from a wide-area network of reference stations via satellite communications, etc., allowing users to apply real-time corrections on their side. A representative example in Japan is QZSS (Michibiki)’s “Centimeter-Class Augmentation Service (CLAS).” CLAS calculates satellite orbit and clock errors and atmospheric errors (ionosphere and troposphere) based on data from government-operated CORS networks and broadcasts that correction information nationwide from Michibiki satellites on the L6 band. Users with compatible receivers can receive correction signals directly from satellites even in mountainous areas or at sea where communication coverage is unavailable, enabling centimeter-class positioning solo. This has the major advantages that users do not need to prepare their own base stations and do not require communications infrastructure to receive corrections. Moreover, CLAS is provided free by the government (though compatible equipment must be purchased), making it a groundbreaking initiative that lowers the cost barrier to high-precision positioning.

The PPP-RTK concept itself is being researched and commercialized worldwide; in Europe, high-precision services using Galileo satellites have started, and private companies are also deploying proprietary satellite-based augmentation services (L-band correction signals, etc.). As wide-area correction services diversify further, users will be able to obtain RTK accuracy in the optimal way for their applications and regions. With the development of network RTK and PPP-RTK, the idea of “centimeter accuracy anywhere with ease” is becoming a reality.

Miniaturization of high-precision GNSS equipment and easier use

Receivers and antennas for handling RTK positioning have also undergone miniaturization and cost reduction in recent years. Previously, centimeter-level positioning required expensive surveying equipment or large fixed GNSS installations. Recently, however, inexpensive high-precision GNSS modules and chipsets have appeared on the market, making it not impossible for individuals to obtain centimeter accuracy within reach. For example, small receiver boards that support multi-GNSS and dual-frequency reception can now be obtained for around tens of thousands of yen, and handheld GNSS receivers or smartphone-connected devices incorporating them have increased.

Particularly notable is integration with smartphones. Some Android models and recent iPhones have begun to include dual-frequency GNSS chips necessary for high-precision positioning. Android also provides APIs to obtain raw GNSS data for use in RTK processing, allowing smartphones themselves to be used as rovers. In practice, however, achieving centimeter-level accuracy with a smartphone alone is limited by antenna performance and stability, so many users pair smartphones with small RTK receivers that connect via Bluetooth or USB. By combining these with a smartphone or tablet, tasks that formerly required fixed equipment and surveying poles can now be done with a single handheld device.

The spread of these small, convenient high-precision GNSS devices is making RTK positioning familiar not only to professional surveyors but also to general technicians and end users. Because setup time at worksites is reduced and devices are easy to carry, the efficiency of positioning work has dramatically improved. For example, on construction sites, surveying that previously required two people can now be done by one person, helping alleviate labor shortages and reduce operating costs. As use cases requiring RTK accuracy increase, the trend of miniaturization and ease of use is driving the wider adoption of high-precision positioning technologies.

Further accuracy enhancement through sensor fusion and AI

GNSS-based high-precision positioning heavily depends on satellite signal reception conditions. It is naturally difficult to acquire satellites under building shadows or under trees, and RTK can suffer from interruptions and reduced accuracy. Recently, attention has turned to maintaining and complementing accuracy through fusion with sensors and technologies other than GNSS.

A representative example is integration with inertial measurement units (IMUs). By combining accelerometers and gyros (IMU) with a GNSS receiver, position can continue to be estimated even when GNSS reception is temporarily unavailable. Filtering technologies (e.g., Kalman filters) that seamlessly bridge outages using IMU data inside tunnels or under overpasses and then restore high-precision positions when GNSS is reacquired have been put into practical use. This ensures practical positional accuracy and continuity while minimizing interruptions due to outages.

Another recent trend is the application of artificial intelligence (AI) and machine learning to GNSS positioning. Research is underway to use AI to identify and remove multipath effects (such as building reflections) from received satellite signal data, or to self-learn correction models from past positioning error trends. For example, techniques have been reported that use deep learning to determine whether a satellite signal in an urban area is direct or reflected, removing adverse effects and improving accuracy. Many of these approaches are still at the research stage, but in the future AI could be integrated into GNSS receivers to perform optimal signal processing and sensor fusion in real time according to environmental conditions, creating smart GNSS.

Furthermore, high-precision self-positioning using sensor fusion with cameras and LiDAR (SLAM technologies) is advancing, especially in the autonomous driving field. By incorporating relative environmental information that GNSS alone cannot provide, this approach complements GNSS absolute positioning to enhance overall accuracy. Overall, sensor fusion including IMUs and the use of AI are current trends aimed at stably maintaining and reinforcing RTK’s centimeter accuracy. This will expand the applicability of high-precision positioning and enable reliable position information even in more challenging environments.

New application fields for RTK technology

The evolution of high-precision positioning technologies such as RTK is rapidly expanding their application fields. While surveying and civil measurements were the main uses until now, centimeter-level positioning is currently being applied in a variety of fields, such as:

• Autonomous driving and vehicle navigation: Lane-level position determination on roads requires accuracy on the order of several tens of centimeters or better, so high-precision GNSS plays an important role. Efforts are underway to accurately determine vehicle position in autonomous vehicles and advanced driver assistance systems (ADAS) by combining RTK and augmentation services.

• Agriculture (precision agriculture): Equipping tractors and agricultural machinery with RTK-GNSS enables automatic driving with position control errors of a few centimeters, allowing extremely precise operations (seeding, fertilization, harvesting, etc.). High-precision positioning also supports waste-free operations and night-time autonomous driving.

• Drone surveying and aerial photogrammetry: Mounting high-precision GNSS receivers on drones allows direct geotagging of aerial images. RTK-capable drones can perform photogrammetry at centimeter accuracy without ground control points, greatly improving efficiency in volume calculations and topographic mapping.

• Infrastructure inspection and maintenance: RTK is being used to record photo locations with high precision during inspections of bridges and towers, and for continuous monitoring of small displacements of the ground or structures. High-precision GNSS is effective for detecting changes on the order of centimeters through periodic observations.

• AR navigation and MR technologies: Outdoor AR applications require high-precision self-positioning to correctly place virtual objects in real space. Combining centimeter-class GNSS with compass and IMU enables user positioning outdoors with sub-meter errors, allowing seamless navigation and guidance displays on AR glasses or smartphones.

Thus, RTK’s increased accuracy and usability are spawning new use cases. In Japan, the Ministry of Land, Infrastructure, Transport and Tourism promotes productivity improvements in construction under the initiative called i-Construction, and high-precision GNSS is one of the key technologies. RTK is being introduced on-site to improve efficiency and reduce labor in surveying and as-built control, and demand is expected to grow. Centimeter positioning, once the domain of specialists, is becoming infrastructure used routinely in many aspects of society.

Simple surveying with LRTK

The RTK accuracy trends introduced so far are opening an era in which “anyone can use high-precision positioning anywhere.” A symbolic example of this is a new high-precision GNSS solution called LRTK. LRTK is a platform that combines a small GNSS receiver with a smartphone app and cloud services, aiming to realize simple surveying without specialized knowledge or complex equipment.

Specifically, LRTK’s small receiver can be attached to a helmet or smartphone, and by simply walking around a site it automatically acquires high-precision position data and point cloud data. Traditional tripod setups and leveling are unnecessary, and RTK positioning can be started with one touch. It supports multi-GNSS and dual-frequency reception and is compatible with Michibiki’s CLAS reception and network RTK via Ntrip. Therefore, centimeter-class real-time positioning can be performed anywhere in Japan without preparing a base station.

LRTK was developed by a startup from the Tokyo Institute of Technology and combines the robustness and ease of use required on construction and civil engineering sites. Positioning results can be viewed in real time on a smartphone screen, and acquired data can be uploaded to the cloud for sharing and analysis. This enables significant improvements in efficiency and labor savings in surveying tasks. In fact, there are reported cases where one-person surveying using the LRTK series significantly reduced work time compared to traditional methods.

In this way, LRTK, which condenses the latest technologies, can be said to embody “easy precision positioning for everyone.” It leverages all the trends in RTK accuracy discussed—multi-GNSS and wide-area augmentation, miniaturized modules, smartphone integration, and cloud use—delivering a tool that is immediately useful on site. If readers are interested in improving work efficiency or trying new surveying styles using high-precision positioning, consider exploring simple surveying with LRTK.

FAQ

Q: What is RTK positioning? How much accuracy can it provide? A: RTK positioning is a method that obtains centimeter-level accuracy by receiving GNSS signals simultaneously at a base station and a rover and applying real-time error corrections. Typical accuracy is said to be around 2–3 cm horizontally and 4–5 cm vertically. However, it varies with environmental conditions and baseline length, and under favorable conditions even higher accuracy can be expected.

Q: What is the difference between standalone positioning or ordinary GPS and RTK? A: Standalone positioning (standalone GPS/GNSS) uses only satellite signals and therefore has errors on the order of meters. In contrast, RTK uses correction data from a reference station to cancel satellite signal error sources and compute a high-precision position. In short, if ordinary GPS tells you “approximately several meters,” RTK is a technology that can determine position to “within a few centimeters.”

Q: What are the benefits of multi-GNSS and dual-frequency support? A: Multi-GNSS support increases the number of available satellites, making it easier to secure the satellites needed for positioning even in building shadows or mountainous areas. Dual-frequency (multi-frequency) support enables removal of ionospheric errors and rapid integer ambiguity resolution, shortening initial convergence time and stabilizing accuracy. Simply put, the benefit is “using many satellites’ information and removing radio errors to achieve more accurate and less interrupted positioning.”

Q: What are PPP-RTK and CLAS? How do they differ from traditional RTK? A: PPP-RTK distributes wide-area error information via satellite communications, etc., allowing real-time corrections to be applied by the user. Japan’s CLAS (Michibiki’s Centimeter-Class Augmentation Service) is a type of PPP-RTK, where correction information derived from government-operated reference networks is provided free via satellite. Traditional RTK required users to set up their own base stations or receive corrections via communications, but PPP-RTK/CLAS eliminates that need. The major difference is that high-precision positioning is possible even outside communication coverage or where no base stations are installed.

Q: Is RTK positioning possible with a smartphone? A: Some recent smartphones include dual-frequency GNSS, and in theory a smartphone alone can perform RTK positioning. However, due to antenna performance and stability, practical standalone use is difficult, so in reality smartphones are often used together with external RTK-capable receivers that connect via Bluetooth. With such small devices (e.g., LRTK), a smartphone can serve as the display and controller to conveniently utilize centimeter-accuracy positioning.

Q: What is LRTK? A: LRTK is a solution consisting of a small high-precision GNSS receiver and a dedicated app, designed to enable anyone to perform centimeter-class positioning and surveying easily. It works in conjunction with smartphones and supports Michibiki CLAS and network RTK, providing high accuracy without a base station. Compared with traditional surveying equipment, it features simpler setup and operation, enabling efficient on-site positioning and measurement by a single person.