What is RTK? Real-Time Kinematic (RTK) GNSS Positioning Explained – Accuracy

RTK (Real-Time Kinematic) is a GNSS positioning technique that delivers centimeter-level accuracy in real time — roughly 100× more precise than standard GPS. It works by using a base station at a known location to compute satellite signal errors and transmit corrections to a mobile receiver (rover), canceling out atmospheric delays, orbital errors, and clock drift that limit ordinary GPS to 10–30 ft (3–10 m) of accuracy.

This guide covers how RTK works, the science behind its precision, how it compares to other correction methods (DGNSS, PPK, PPP), the equipment and infrastructure you need to get started, real-world applications across industries, and answers to the most common questions. We also introduce LRTK, a smartphone-based RTK + AR solution that is making centimeter-level positioning accessible to anyone on a construction site.

Table of Contents

• How RTK Works: Base Station, Rover, and Carrier Phase

• RTK Solution States: Fix, Float, and Single

• Error Sources in GPS and How RTK Corrects Them

• RTK Accuracy: What to Realistically Expect

• RTK vs DGNSS vs PPK vs PPP: Choosing the Right Method

• Types of RTK Systems: Own Base, Network RTK, CORS, and VRS

• Equipment and Costs: What You Need to Get Started

• RTK Applications Across Industries

• The Rise of Smartphone RTK

• AR × RTK: Bringing Design Data Into the Real World

• LRTK: Smartphone RTK + AR for Construction

• FAQ

How RTK Works: Base Station, Rover, and Carrier Phase

At its core, RTK uses two GNSS receivers working together: a base station (set at a known, precisely surveyed location) and a rover (the mobile unit whose position you want to determine). Both receivers track signals from the same GPS, GLONASS, Galileo, and BeiDou satellites simultaneously. Because the base station already knows its exact coordinates, it can calculate how much the satellite signals are being distorted and send that information — called correction data — to the rover in real time. The rover applies those corrections to its own measurements, effectively canceling out the errors that make standard GPS inaccurate.

Why Carrier Phase Matters

The key to RTK's precision lies in carrier phase measurements. Standard GPS positioning relies on the navigation message encoded in the satellite signal (called code-phase or pseudorange measurement), which has a resolution on the order of meters. RTK takes a fundamentally different approach: it measures the phase of the carrier wave itself.

The GPS L1 carrier signal has a wavelength of approximately 7.5 in (19 cm). A receiver can measure where it sits within a single wavelength to sub-millimeter precision — but the signal repeats every 19 cm, so the receiver doesn't initially know the total number of complete wavelengths between the satellite and itself. This is called the integer ambiguity problem.

The distance to the satellite can be expressed as:

where N is the unknown whole number of cycles. RTK receivers solve this ambiguity by evaluating all possible integer combinations across multiple satellites and frequencies simultaneously, using algorithms such as the LAMBDA method (Least-squares AMBiguity Decorrelation Adjustment). When the receiver is confident it has found the correct combination, it declares a Fix solution and unlocks full centimeter-level precision.

More satellites and more frequencies provide more independent equations, which is why multi-constellation, multi-frequency receivers resolve ambiguities faster and more reliably.

RTK Solution States: Fix, Float, and Single

When working with RTK, your receiver will report one of three solution states. Understanding these is critical to knowing whether your data is trustworthy:

Single — No correction data is being received or applied. The receiver is operating like a standard GPS, with accuracy of roughly 3–10 ft (1–3 m). Data collected in Single mode is not suitable for precision work.

Float — Corrections are being received and applied, but the integer ambiguities have not yet been fully resolved. Accuracy is typically 8–20 in (20–50 cm). Float solutions are a transitional state — better than Single, but not yet at centimeter precision.

Fix — The integer ambiguities have been resolved with high statistical confidence. This is the target state for all precision work, delivering accuracy of roughly 0.5–1 in (1–2 cm) horizontally. Always confirm your receiver is in Fix before collecting critical measurements.

The time it takes to go from Single to Fix (called initialization or time to first fix) depends on satellite visibility, the number of tracked constellations, signal quality, and the baseline length. Under good conditions with a modern multi-frequency receiver, initialization typically takes 10–60 seconds.

Error Sources in GPS and How RTK Corrects Them

Standalone GPS is affected by multiple error sources that collectively cause position deviations of 3–30 ft (1–10 m). RTK corrects most of these by comparing the base station's known position with what the satellites report:

Satellite Orbit Errors — Satellites don't follow their predicted orbits perfectly. Small deviations translate into position errors on the ground. Because the base station and rover see the satellites from nearly the same angle, this error cancels in the differential calculation.

Satellite Clock Errors — Even atomic clocks on satellites drift slightly. Again, because both receivers observe the same clock error at nearly the same time, the difference cancels out.

Ionospheric Delay — The ionosphere (50–600 miles above Earth) slows and bends GNSS signals, adding variable delays. When the base station and rover are relatively close together (within about 6–12 mi / 10–20 km), they experience nearly identical ionospheric conditions, so the error cancels. At longer baselines, this is the dominant source of residual error.

Tropospheric Delay — The lower atmosphere (weather, humidity, temperature) also delays signals. Like ionospheric delay, this cancels well at short baselines.

Multipath — Signals bouncing off buildings, vehicles, rock faces, or the ground before reaching the antenna cause "echoes" that distort measurements. Multipath is local to each receiver and cannot be fully corrected by RTK. Mitigation strategies include using high-quality antennas with ground planes, choosing open-sky locations, and raising the antenna height.

Receiver Noise — Internal electronic noise in the receiver. This is generally very small in modern equipment and cannot be corrected remotely.

The bottom line: RTK effectively eliminates satellite-related and atmospheric errors, but cannot fix multipath or receiver noise. This is why site selection and antenna quality still matter, even with RTK.

RTK Accuracy: What to Realistically Expect

Under good conditions (open sky, short baseline, multi-frequency receiver, Fix solution), RTK delivers:

• Horizontal accuracy: ±0.4–0.8 in (±1–2 cm) + 1 ppm of baseline distance

• Vertical accuracy: ±0.6–1.2 in (±1.5–3 cm) + 1 ppm of baseline distance

The "1 ppm" term means accuracy degrades by about 0.04 in (1 mm) for every 0.6 mi (1 km) of baseline. At 6 mi (10 km), horizontal error becomes roughly ±0.8 in + 0.4 in = ±1.2 in (±3 cm). At 12 mi (20 km), RTK approaches its practical limit.

Vertical accuracy is always worse than horizontal — typically 1.5–2× larger. This is a fundamental consequence of satellite geometry: GNSS satellites are above you, never below, so the vertical component has weaker geometric strength.

Factors that degrade accuracy:

• Obstructed sky view (urban canyons, tree canopy, steep terrain)

• Long baseline distance to the nearest reference station

• High ionospheric activity (solar storms, equatorial regions)

• Multipath-heavy environments (near large metal structures or water)

• Float solution instead of Fix

RTK vs DGNSS vs PPK vs PPP: Choosing the Right Method

RTK is one of several GNSS correction methods. Each has trade-offs in accuracy, real-time capability, infrastructure needs, and cost:

RTK vs PPK: Both achieve centimeter accuracy using carrier phase, but RTK corrects in real time while PPK processes corrections after the fact. RTK requires a live data link to the base station; PPK does not. PPK is more robust when communication links are unreliable (e.g., drone flights over mountains), while RTK gives you instant feedback — essential for stakeout and machine guidance.

RTK vs PPP: PPP (Precise Point Positioning) uses globally broadcast satellite correction data, so it works anywhere without a local base station. However, PPP requires a convergence period of several minutes to tens of minutes to reach full accuracy, and its precision is typically decimeter-level rather than centimeter-level, although newer PPP-RTK hybrid services are closing this gap.

RTK vs DGNSS: Both use a base station and rover, but DGNSS relies on code-phase corrections (meter-level) while RTK uses carrier-phase corrections (centimeter-level). DGNSS is simpler and works over longer baselines, but an order of magnitude less precise.

How to choose: If you need centimeter accuracy right now in the field (stakeout, machine control, live surveying), use RTK. If you can wait for post-processing and need reliability over communication, use PPK. If no local base is available, consider PPP. If sub-meter is good enough, DGNSS or SBAS may suffice at lower cost.

Types of RTK Systems: Own Base, Network RTK, CORS, and VRS

There are two fundamental ways to get correction data for RTK: set up your own base station, or subscribe to a reference station network.

Own-Base RTK

You set up a GNSS receiver at a known point near your work site and transmit corrections to your rover via radio (UHF, typically 400–470 MHz) or Bluetooth/Wi-Fi for short range. This approach gives you full control and works in areas without cell coverage.

• Pros: No subscription fees, works offline, full control

• Cons: Requires two receivers, base must be precisely located, effective range limited to roughly 6 mi (10 km), accuracy degrades with distance

Network RTK (CORS and VRS)

A CORS (Continuously Operating Reference Station) network consists of permanently installed GNSS base stations that stream correction data over the internet. In the United States, several options are available:

• NGS CORS Network: Over 2,000 stations managed by NOAA's National Geodetic Survey. Primarily for post-processing (OPUS), not real-time RTK.

• Private RTK Networks: Services like Trimble VRS Now (CenterPoint RTX), SmartNet North America, KeyNetGPS, and SwiftNav Skylark operate dense networks of reference stations and sell real-time RTK subscriptions covering most of the continental U.S.

• State DOT networks: Some states (e.g., Ohio, North Carolina, Oregon) operate free or low-cost real-time CORS networks.

VRS (Virtual Reference Station) is a technique used by network RTK providers. Instead of sending corrections from a single physical station, the network calculates what a hypothetical base station right next to you would observe, based on data from the surrounding real stations. This gives you short-baseline accuracy regardless of your distance to any single station.

Corrections from CORS/VRS networks are delivered to your rover via the NTRIP (Networked Transport of RTCM via Internet Protocol) standard. Your rover connects to the NTRIP caster over cellular data, and corrections flow in real time.

• Pros: No base station setup, one-person operation, consistent accuracy over large areas

• Cons: Requires cell coverage, subscription fees ($50–$200+/month), dependent on network uptime

Equipment and Costs: What You Need to Get Started

The cost of entry into RTK has dropped dramatically in recent years:

RTK Receivers

• Budget hobbyist/maker boards (e.g., u-blox ZED-F9P based): $200–$500

• Mid-range surveying rovers: $2,000–$8,000

• Professional multi-frequency receivers (Trimble, Leica, Topcon): $8,000–$25,000+

• Smartphone RTK receivers (compact Bluetooth units): $500–$3,000

Base Station (if setting up your own)

• Same receiver as above, plus antenna, tripod, and radio modem: $2,000–$15,000+

Network RTK Subscription (if using CORS/VRS)

• Typical range: $50–$200+ per month per device

• Some state networks are free

Software

• Many receivers come with field software; open-source options like RTKLIB are available for post-processing

• Commercial survey/GIS apps: varies widely

Total cost to get started: As low as a few hundred dollars for a DIY setup with a budget receiver, or $3,000–$10,000 for a professional single-rover system using network RTK.

RTK Applications Across Industries

RTK's centimeter-level, real-time accuracy has made it essential in a growing number of fields:

Surveying and Mapping

The original and still most common RTK application. Land surveyors use RTK rovers for boundary surveys, topographic mapping, control point establishment, and construction stakeout. Tasks that once required hours with a total station can often be completed in minutes with RTK.

Construction and Civil Engineering

Under initiatives like ICT-driven construction, RTK is integral to machine guidance and machine control. Bulldozers, excavators, and graders equipped with RTK-GNSS receivers can follow design grades automatically, reducing the need for stakes and string lines. RTK is also used for as-built verification, earthwork volume calculations, and quality control.

Precision Agriculture

RTK enables auto-steer systems on tractors and combines, allowing pass-to-pass accuracy of under 1 in (2 cm). This reduces overlap, saves fuel and inputs (seed, fertilizer, pesticide), and enables variable-rate application. According to USDA estimates, precision agriculture technologies can reduce input costs by 10–20%.

Drone and UAV Operations

RTK (and PPK) enable drones to capture aerial imagery with centimeter-accurate geotags, reducing or eliminating the need for ground control points in photogrammetry. This is widely used for stockpile measurement, mine monitoring, corridor mapping (pipelines, power lines, roads), and environmental surveys.

Autonomous Vehicles and Robotics

Self-driving cars, delivery robots, and autonomous mining equipment rely on RTK for lane-level or better positioning. RTK's real-time nature makes it suitable for safety-critical navigation where delays are unacceptable.

Infrastructure Inspection and Maintenance

Bridge deflection monitoring, railway track inspection, dam deformation surveys, and utility asset mapping all benefit from RTK's precision. By georeferencing inspection data, organizations can track changes over time with millimeter-level sensitivity.

The Rise of Smartphone RTK

Until recently, RTK required expensive, dedicated hardware — bulky receivers, external antennas, and specialized data collectors. That's changing fast.

Modern smartphones now contain multi-band GNSS chips (e.g., Qualcomm's platforms support L1+L5), and compact external RTK receivers that pair via Bluetooth can upgrade a standard phone to centimeter-level accuracy. These pocket-sized receivers weigh as little as 4–5 oz (110–140 g) and can connect to NTRIP correction services or receive satellite-based PPP-RTK corrections directly.

The implications are significant:

• Lower cost of entry: A smartphone + compact RTK receiver costs a fraction of traditional surveying equipment.

• One device per person: Every crew member can carry a sub-inch positioning tool in their pocket.

• Instant data integration: Survey points, photos, and notes can be uploaded to the cloud immediately.

• App-based workflows: Intuitive apps replace clunky data collectors, reducing training time.

Smartphone RTK is not a replacement for professional survey-grade equipment in all scenarios — but for construction layout, asset mapping, as-built checks, and many field GIS tasks, it is increasingly sufficient.

AR × RTK: Bringing Design Data Into the Real World

One of the most exciting developments in the RTK space is the combination of high-precision positioning with AR (Augmented Reality). When a smartphone knows its position to within an inch, it can overlay 3D design models, CAD drawings, and BIM data onto the real world through the camera view — and have them stay locked in place as the user moves around.

Traditional AR on construction sites suffered from drift: virtual models would shift as the user walked. RTK solves this by anchoring the AR display to absolute coordinates, not just visual features. The result is a coordinate-linked AR experience that can:

• Show where to drive stakes without physical layout marking

• Overlay planned excavation lines, foundation outlines, or pipe routes on the ground

• Visualize underground utilities before digging

• Compare as-built conditions against the 3D design model in real time

• Guide workers to exact positions on site with virtual navigation arrows

This fusion of RTK and AR is transforming how field crews interact with design data — making it visible, spatial, and immediate rather than confined to paper plans or a laptop in the trailer.

LRTK: Smartphone RTK + AR for Construction

LRTK is a solution that brings the power of RTK + AR to any construction site through a smartphone. It consists of a palm-sized RTK-GNSS receiver (LRTK Phone) and a dedicated app, and it is designed to make centimeter-level positioning and AR visualization accessible to anyone — not just survey professionals.

How It Works

Attach the compact receiver (about 4.4 oz / 125 g, 0.5 in / 13 mm thick) magnetically to your iPhone or iPad, launch the app, and you're positioned with sub-inch accuracy. LRTK acquires corrections from network RTK services via NTRIP or from satellite-delivered PPP-RTK signals, so it works with or without your own base station.

Key Features

• One-tap point capture: Record latitude, longitude, and elevation with a single tap. Coordinates are automatically converted to standard systems (State Plane, UTM, WGS84).

• AR overlay mode: Load CAD, BIM, or design data into the app and see it superimposed on the real world through your camera. Models stay locked to absolute coordinates — no drift.

• Tilt compensation: Some models correct for pole tilt, so you get accurate ground-point coordinates even when the pole isn't perfectly vertical.

• Offline positioning: Models with satellite-based correction support continue working outside cellular coverage — useful for remote mountain roads, offshore platforms, or rural farmland.

• Cloud sync: Measured points, photos, and notes upload instantly to a cloud map platform where office staff and remote teams can view them in real time.

• Built-in calculations: Distance, area, and volume tools let you do earthwork quantity checks on the spot.

Who Uses LRTK

LRTK is deployed across a range of field applications:

• Pile driving guidance — Import pile coordinates and see virtual markers where each pile should go.

• As-built verification — Overlay the design model on the finished earthwork and instantly see discrepancies.

• Stakeout / layout — Follow on-screen AR markers to mark positions, replacing traditional batter boards and string lines.

• Boundary marker location — Pre-register known coordinates and locate stakes hidden by vegetation using AR.

• Infrastructure inspection — Georeference crack locations, displacement measurements, and inspection photos for centralized tracking.

• Disaster response — In scenarios where cell networks fail, satellite-corrected models continue working for rapid damage assessment and georeferenced documentation.

Benefits of Implementation

Teams that have adopted LRTK report significant gains:

• Time savings: Workflows that took days (plan → survey → verify → report) collapse into same-day cycles. One crew completed control point surveying through as-built checks in half a day with a single smartphone.

• Cost reduction: A smartphone + LRTK receiver replaces multiple pieces of expensive equipment. In-house teams can handle tasks previously outsourced to survey contractors.

• Reduced crew size: Tasks that required 2–3 people (instrument operator + rod person) can now be done by one.

• Safety: AR visualization of buried utilities before excavation prevents accidental strikes. Fewer people on site means less exposure to heavy equipment hazards.

• Faster onboarding: Intuitive, app-based interface means new crew members get productive in hours rather than weeks of instrument training.

Because LRTK runs on the smartphones people already carry, there's no steep learning curve and no bulky equipment to transport. It is designed to lower the barrier to high-precision positioning for contractors and field crews of any size.

FAQ

What is the difference between RTK and regular GPS?

Regular GPS (standalone GNSS) uses a single receiver to calculate position from satellite signals. Because it cannot correct for atmospheric delays, orbital errors, or clock drift, accuracy is limited to roughly 3–30 ft (1–10 m). RTK uses two receivers — a base station at a known location and a rover — so the base can compute the exact errors affecting the satellite signals and transmit corrections to the rover in real time. By canceling these shared errors, RTK reduces positioning error from meters to about 0.5–1 in (1–2 cm). The key technical difference is that RTK analyzes the carrier phase of satellite signals (wavelength ~7.5 in / 19 cm), while standard GPS relies on the navigation code (wavelength ~1,000 ft / 300 m), giving RTK roughly 100× better measurement resolution.

What equipment is required for RTK positioning?

At minimum, you need an RTK-capable GNSS receiver (rover) and a source of correction data. For the correction source, you can either set up your own base station (a second GNSS receiver at a known point, plus a radio or internet link) or subscribe to a network RTK service that streams corrections via the internet using the NTRIP protocol. If you use a network service, all you need in the field is the rover, a cellular connection, and an NTRIP subscription. Modern compact receivers that pair with smartphones via Bluetooth (such as LRTK) make the setup even simpler — no specialized data collector required.

What level of accuracy can RTK achieve?

Under ideal conditions (open sky, short baseline, multi-frequency receiver, Fix solution), RTK achieves horizontal accuracy of about ±0.4–0.8 in (±1–2 cm) and vertical accuracy of about ±0.6–1.2 in (±1.5–3 cm). Accuracy degrades by approximately 1 mm per km (1 ppm) of baseline distance. In practice, factors like obstructed sky view, multipath, long baselines, and high ionospheric activity can reduce accuracy. It is also important to note that vertical accuracy is always 1.5–2× worse than horizontal due to satellite geometry.

How far can the rover be from the base station?

For traditional single-base RTK, the practical limit is about 6–12 mi (10–20 km). Beyond that, differences in atmospheric conditions between the base and rover become too large to cancel, and the receiver may struggle to maintain a Fix solution. Network RTK (VRS) effectively solves this problem by creating a virtual base station near the rover's location, regardless of the physical distance to any single reference station. With a good network RTK service, you can maintain centimeter accuracy across an entire region.

What is the difference between RTK and PPK?

Both RTK and PPK use carrier-phase measurements to achieve centimeter-level accuracy. The difference is when corrections are applied. RTK applies corrections in real time over a live data link, giving you instant results in the field. PPK records raw GNSS data from both the base and rover, and corrections are applied after the fact on a computer. RTK is essential when you need immediate feedback (stakeout, machine guidance). PPK is more robust for applications where the data link may be unreliable (drone flights, remote areas) and allows forward-backward processing that can fill in gaps.

Is RTK used in fields other than surveying?

Absolutely. RTK is widely used in precision agriculture (auto-steer tractors, variable-rate application), drone mapping and photogrammetry, construction machine guidance and control, autonomous vehicle navigation, infrastructure inspection, railway track maintenance, and even robotics. Any application where real-time centimeter-level positioning adds value is a potential RTK use case.

Can RTK be used with a smartphone?

Yes. By pairing a compact, external RTK-GNSS receiver (connected via Bluetooth) with a smartphone, you can achieve sub-inch accuracy on a device that fits in your pocket. Solutions like LRTK combine a lightweight receiver with a dedicated app that handles NTRIP connections, coordinate conversion, data logging, and even AR visualization — turning your phone into a capable field surveying tool. While smartphone RTK may not replace survey-grade instruments for the most demanding control-network work, it is increasingly sufficient for construction layout, asset mapping, as-built checks, and many field GIS applications.

Are RTK-compatible devices expensive?

The cost has dropped significantly. Budget RTK receiver boards (e.g., based on the u-blox ZED-F9P) start under $300. Mid-range survey rovers run $2,000–$8,000. Professional multi-frequency systems from major manufacturers can exceed $15,000. Smartphone-based RTK receivers like LRTK fall in the lower end of the range, and because they use your existing smartphone as the display and data collector, the total system cost is far lower than a traditional setup. Network RTK subscriptions typically add $50–$200/month. Overall, the barrier to entry for centimeter-accurate positioning is lower than it has ever been.

Does RTK work in environments with poor satellite visibility, like urban canyons or forests?

RTK accuracy depends heavily on satellite signal quality. In urban canyons (surrounded by tall buildings), dense forests, or near large metal structures, signals can be blocked or reflected (multipath), making it difficult to achieve or maintain a Fix solution. Using a multi-GNSS, multi-frequency receiver helps by increasing the number of trackable satellites — even if some are blocked, others may be visible. Strategies like temporarily moving to a more open location for initialization, raising the antenna higher, or using an antenna with a ground plane to reduce multipath can also help. However, in environments where the sky is almost entirely obstructed (inside tunnels, underground, indoors), RTK cannot work and alternative technologies (total stations, IMUs, SLAM) must be used instead.

What is NTRIP and do I need it for RTK?

NTRIP (Networked Transport of RTCM via Internet Protocol) is the standard protocol for streaming RTK correction data over the internet. If you are using a network RTK service (CORS, VRS) rather than your own local base station and radio link, then yes — NTRIP is how your rover receives corrections. To use NTRIP, you need a cellular data connection on your rover (or phone) and credentials from a correction service provider: a host address, port number, mountpoint, and username/password. Many RTK apps handle NTRIP configuration with a simple settings screen.