Skip to content

Real-Time Kinematic Positioning

A relative GNSS positioning method that combines simultaneous rover and known-reference carrier-phase observations, real-time correction transport, and validated integer-ambiguity resolution to produce high-precision coordinates in the reference station's frame.

Version
v1 · 2026-08-30 · History
Domain-specific #
2635
Origin domain
geodetic surveying
Subdomain
carrier-phase differential GNSS
Aliases
RTK positioning, GNSS RTK

Core Idea

Real-time kinematic positioning (RTK) is a relative Global Navigation Satellite System (GNSS) method that estimates a moving or stationary rover antenna's coordinates from carrier-phase and code observations made simultaneously at the rover and at a reference receiver whose coordinates are known. Reference observations or derived corrections reach the rover through a low-latency data link; the rover estimates the baseline and, when evidence supports it, resolves carrier-phase ambiguities to integers. A validated fixed solution can deliver centimetre-class relative positioning without waiting for post-processing.

Scope of Application

The abstraction applies to single-base RTK and network RTK when the network synthesizes corrections appropriate to a rover location. Classical single-base RTK streams base observations or corrections to one or more rovers. Network RTK uses several reference stations to estimate spatially varying errors and may deliver a virtual-reference-station stream, master–auxiliary information, or another correction representation. The network changes how reference information is produced; it does not remove the rover, carrier-phase, real-time, integer-resolution, coordinate-frame, or validation roles.

Clarity

A practical diagnostic asks five questions: Where is the reference? What simultaneous carrier observations or corrections reach the rover? What integer ambiguities were fixed and validated? In which frame and epoch are coordinates expressed? What independent evidence shows the result met its accuracy requirement? If any answer is absent, “RTK” may be a marketing or display label rather than a demonstrated method.

Manages Complexity

RTK compresses a large estimation problem into a reusable operational architecture. Each satellite signal includes clocks, orbit, atmosphere, antenna, receiver, multipath, code, phase, and cycle-continuity effects. A nearby known receiver converts many of those effects from unknowable absolute errors into shared terms that can be cancelled, differenced, or estimated. Integer ambiguity resolution converts very precise phase change into an absolute relative range within each tracking arc. The correction link makes that inference available soon enough to guide a surveyor or control a machine.

Abstract Reasoning

RTK licenses several disciplined inferences. First, benefits from differencing depend on error correlation, not error magnitude alone. A large satellite-clock term common to base and rover can cancel; smaller local multipath that differs between antennas may remain. Second, precision improves sharply after correct ambiguity fixing because the estimator restricts a continuous uncertainty region to an integer lattice. Incorrect restriction can create a correspondingly sharp wrong answer, so validation must accompany precision.

Knowledge Transfer

Within geodesy, the signature transfers directly across constellations, carrier bands, correction encodings, radios, cellular links, base brands, and application platforms. The identity survives because it is defined by reference-relative simultaneous carrier observations, ambiguity resolution, real-time delivery, and a frame-tied coordinate—not by GPS L1, UHF, one vendor, or one message version. RTCM's differential-GNSS standards and Ntrip transport support interoperability, while equipment-specific processing remains variable.

Relationships to Other Abstractions

Local relationship map for Real-Time Kinematic PositioningParents appear above the current abstraction, mutual partners to the right, and children below. Node labels state whether each abstraction is prime or domain-specific; colors identify relation types.Real-Time KinematicPositioningDOMAINPrime abstraction: Measurement — is a kind ofMeasurementPRIME

Current abstraction Real-Time Kinematic Positioning Domain-specific

Parents (1) — more general patterns this builds on

  • Real-Time Kinematic Positioning is a kind of Measurement Prime

    The minimal proposed parent is prime:measurement.

Hierarchy path (1) — routes to 1 parentless root

Neighborhood in Abstraction Space

Real-Time Kinematic Positioning sits in a sparse region of the domain-specific corpus (90th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (1565 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-09-08