1 Hz · RTK
GUIDE-K35C
1 Hz RTK configuration without inertial dead reckoning, evaluated at module level on a development board.
View GUIDE product page →Ground Truth Field Test · 02
K16 and K35 GNSS Ground Truth Field Test: Multi-Configuration Vehicle Route Performance
This Ground Truth field test evaluates four Kalmix GNSS module configurations over a common vehicle-route analysis window. All four samples use the same car platform and full-band survey-grade antenna setup, with measured positioning performance assessed against the same ground-truth reference across Highway, Tunnel, Residential, and Urban Canyon environments.
Test Overview
Test conditions & dataset
Basic Test Information
Test time
2026-07-03
Test platform
Car
Antenna
Full-band survey-grade antenna
Test samples
GUIDE-K35C/E, NOMAD-K16C/E
Dataset Summary
Output rates
1 / 10
Hz
Epochs
55,601
Total distance
50.4
km
Duration
01:32:40
Route Overview
Test Samples
What was tested
All four samples were evaluated under the same vehicle, antenna, power, correction, and reference conditions.
10 Hz · RTK + DR
GUIDE-K35E
10 Hz RTK configuration with inertial dead reckoning, evaluated at module level on a development board.
View GUIDE product page →
1 Hz · RTK
NOMAD-K16C
1 Hz RTK configuration without inertial dead reckoning, evaluated at module level on a development board.
Coming Soon
10 Hz · RTK + DR
NOMAD-K16E
10 Hz RTK configuration with inertial dead reckoning, evaluated at module level on a development board.
Coming SoonFull-run Results
Measured full-route performance
Select a tested configuration to review full-run accuracy and solution-quality distribution over the common analysis window.
Full-run result
GUIDE-K35C
A 1 Hz RTK module configuration without inertial dead reckoning, evaluated across the common full-route analysis window.
Horizontal RMS
1.400
m
Vertical RMS
1.964
m
CEP50
0.0097
m
R95
3.399
m
Analyzed epochs
5,329
RTK Fixed
81.334%
Overall <10 cm
81.316%
Fixed <10 cm
81.226%
Data completeness
95.828%
Position Quality & Accuracy Distribution
Two aligned cumulative bars over theoretical epochs. The light-grey tail marks epochs that were unavailable for the comparison; matching colours connect each accuracy band to its corresponding position-quality tier.
Position Quality
- RTK Fixed4,523 ep81.334%
- RTK Float799 ep14.368%
- DGNSS7 ep0.126%
- Single0 ep0%
- DR0 ep0%
Accuracy Range
- <10 cm4,522 ep81.316%
- 10 cm–1 m101 ep1.816%
- 1–3 m341 ep6.132%
- 3–10 m361 ep6.492%
- >10 m4 ep0.072%
Full-run result
GUIDE-K35E
A 10 Hz RTK configuration with inertial dead reckoning, evaluated across the common full-route analysis window.
Horizontal RMS
4.209
m
Vertical RMS
0.666
m
CEP50
0.0193
m
R95
9.252
m
Analyzed epochs
55,601
RTK Fixed
80.387%
Overall <10 cm
77.866%
Fixed <10 cm
77.410%
Data completeness
100.000%
Position Quality & Accuracy Distribution
Two aligned cumulative bars over theoretical epochs. The light-grey tail marks epochs that were unavailable for the comparison; matching colours connect each accuracy band to its corresponding position-quality tier.
Position Quality
- RTK Fixed44,696 ep80.387%
- RTK Float8,422 ep15.147%
- DGNSS0 ep0%
- Single151 ep0.272%
- DR2,332 ep4.194%
Accuracy Range
- <10 cm43,294 ep77.866%
- 10 cm–1 m3,698 ep6.651%
- 1–3 m3,166 ep5.694%
- 3–10 m4,107 ep7.387%
- >10 m1,336 ep2.403%
Full-run result
NOMAD-K16C
A 1 Hz RTK configuration without inertial dead reckoning, evaluated across the common full-route analysis window.
Horizontal RMS
2.046
m
Vertical RMS
0.490
m
CEP50
0.0203
m
R95
2.822
m
Analyzed epochs
5,340
RTK Fixed
77.108%
Overall <10 cm
74.213%
Fixed <10 cm
74.015%
Data completeness
96.026%
Position Quality & Accuracy Distribution
Two aligned cumulative bars over theoretical epochs. The light-grey tail marks epochs that were unavailable for the comparison; matching colours connect each accuracy band to its corresponding position-quality tier.
Position Quality
- RTK Fixed4,288 ep77.108%
- RTK Float981 ep17.641%
- DGNSS50 ep0.899%
- Single10 ep0.180%
- DR0 ep0%
Accuracy Range
- <10 cm4,127 ep74.213%
- 10 cm–1 m635 ep11.419%
- 1–3 m330 ep5.934%
- 3–10 m231 ep4.154%
- >10 m6 ep0.108%
Full-run result
NOMAD-K16E
A 10 Hz RTK configuration with inertial dead reckoning, evaluated across the common full-route analysis window.
Horizontal RMS
21.145
m
Vertical RMS
0.963
m
CEP50
0.0173
m
R95
2.808
m
Analyzed epochs
55,601
RTK Fixed
81.407%
Overall <10 cm
71.256%
Fixed <10 cm
71.229%
Data completeness
100.000%
Position Quality & Accuracy Distribution
Two aligned cumulative bars over theoretical epochs. The light-grey tail marks epochs that were unavailable for the comparison; matching colours connect each accuracy band to its corresponding position-quality tier.
Position Quality
- RTK Fixed45,263 ep81.407%
- RTK Float7,086 ep12.744%
- DGNSS834 ep1.500%
- Single64 ep0.115%
- DR2,354 ep4.234%
Accuracy Range
- <10 cm39,619 ep71.256%
- 10 cm–1 m10,093 ep18.153%
- 1–3 m3,463 ep6.228%
- 3–10 m509 ep0.915%
- >10 m1,917 ep3.448%
Overall Error Timeline
Compare the horizontal and vertical error of all four tested configurations across the common analysis window. Hover to inspect each sample at the same elapsed time.
NOMAD-K16E
GUIDE-K35E
NOMAD-K16C
GUIDE-K35C
Loading time-series data…
Scenario Results
Per-environment performance
Each scenario shows the desensitized trajectory video with in-video error readout, plus the same standardized statistics.
Accuracy Range
<10 cm
10 cm–1 m
1–3 m
3–10 m
>10 m
Position Quality
RTK Fixed
RTK Float
DGNSS
Single
DR
01
Highway
Open-road positioning performance across the common four-sample test setup.
- Duration
- 00:12:24
- Distance
- 17.39 km
- Reference epochs
- 7,441
0:00 / 0:00
Core Accuracy Metrics
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
02
Residential
A low-rise residential environment with open-sky visibility between local obstructions.
- Duration
- 00:10:58
- Distance
- 2.83 km
- Reference epochs
- 6,581
0:00 / 0:00
Core Accuracy Metrics
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
03
Urban Canyon
Dense high-rise multipath conditions across the common four-sample test setup.
- Duration
- 00:21:20
- Distance
- 3.51 km
- Reference epochs
- 12,801
0:00 / 0:00
Core Accuracy Metrics
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
04
Tunnel
Inertial dead-reckoning continuity for E configurations; no comparable C-configuration statistics.
- Duration
- 00:02:20
- Distance
- 2.49 km
- Reference epochs
- 1,401
0:00 / 0:00
Core Accuracy Metrics
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
Test Notes
Reading this test
Across the four samples, the open-road segment establishes a stable dual-band RTK baseline.
RTK Fixed rates range from 98.5% to 99.6%, while the reported <10 cm share ranges from 77.9% to 99.7%, confirming consistently usable positioning output under open-sky conditions.
In the low-rise residential segment, performance remains close to the open-road baseline despite intermittent tree cover and local obstructions.
All four samples retain 100% RTK Fixed, showing that dual-band RTK remains stable in this moderate-obstruction environment.
The urban-canyon segment reveals clearer differences between configurations under dense multipath and changing satellite visibility.
This scenario primarily tests RTK algorithm behavior, multipath handling, and recovery strategy.
In the GNSS-denied tunnel, IMU-enabled configurations can maintain stable position output.
Their actual performance still depends heavily on IMU selection and algorithm capability; NOMAD K16 remains under final optimization and validation ahead of its upcoming release.
Explore the Platform