RTK

The RTK HAT (NEO-F9P) supports centimeter-level positioning using RTCM3 corrections.

Note

HAT Compatibility — RTK is available only on the L1/L5 RTK HAT (NEO-F9P).

How RTK Works

Base Station (known position) → RTCM3 corrections → Rover → cm-level fix

Base station determines its precise position (Survey-In or Fixed Position)
Base generates RTCM3 correction messages based on satellite observations
Corrections are sent to the rover (via NTRIP, serial, network, etc.)
Rover applies corrections to achieve centimeter-level accuracy

Fix Quality Progression

Fix Quality	Accuracy	Description
No Fix	—	No satellite signal
GPS 2D/3D	~2–5 m	Standard GNSS fix
Float RTK	~0.2–1 m	Corrections applied, ambiguities not resolved
Fixed RTK	~0.01–0.02 m	Full centimeter-level accuracy

Base Station

C++

auto corrections = hat->rtk()->base()->getTinyCorrections();  // compact (M4M)
auto corrections = hat->rtk()->base()->getFullCorrections();  // full (M7M)
auto frame = hat->rtk()->base()->getRtcm3Frame(1077);         // specific message

Python

corrections = hat.rtk_get_tiny_corrections()   # list[bytes]
corrections = hat.rtk_get_full_corrections()    # list[bytes]
frame = hat.rtk_get_rtcm3_frame(1077)           # bytes

Correction Sets

Method	Type	Messages	Size	Use Case
`getTinyCorrections()`	M4M	Compact subset	Smaller	Low-bandwidth links
`getFullCorrections()`	M7M	Full set	Larger	Best accuracy
`getRtcm3Frame(id)`	Single	One message	Varies	Custom selection

Rover

C++

hat->rtk()->rover()->applyCorrections(corrections);

Python

hat.rtk_apply_corrections(corrections)  # list[bytes]

Configuration

See RTK Config for base mode options (Survey-In, Fixed Position ECEF/LLA) and rover setup.