Introduction
The following datasets were created during the smartLoc project. Our intention is to create one starting point for different groups in varying research areas involving GNSS (Global Navigation Satellite System). These groups could be within the automotive sector focusing on Advanced Driver Assistent Systems. They could also be robotics researchers with focus on localization and mapping or work within the field of communication engineering etc.
Our datasets contain GNSS data from two sensors recorded during real-world urban driving scenarios. On the one hand a mass-market receiver is used and on the other hand the ground truth is derived from a highly accurate reference receiver.
You will find the following data on our website:
- Raw data and fix position of a mass-market receiver
- Ground truth data from a reference receiver related to the mass-market receiver
- Ego motion data from CAN
- Four representative scenarios in urban areas with up to 49% of Non-line-of-sight
- Uniformly time scaled
- Raw data from the reference receiver
- Labeled raw data for Line-of-sight and Non-line-of-sight
More information about each dataset and the setup of our sensors is shown on separate pages. We would be happy to know who uses our datasets or to get an information about the achieved results. Do not hesitate to contact us via e-mail or phone.
Changelog
- 2018-03-01: Fixed end time of RXM-RAWX.csv and NAV-POSLLH.csv in Frankfurt Westend Tower data
- 2018-03-01: Fixed entries 30-32 of RXM-RAWX.csv with the scale factor (Berlin and Frankfurt)
- 2018-02-15: Fixed missing detail for NLOS in RXM-RAWX.csv since the last update (Berlin and Frankfurt)
- 2018-02-15: Renamed Height above mean sea level (GT Height) in Height above ellipsoid (GT Height) in NAV-POSLLH.csv and RXM-RAWX.csv (also for GT Height Cov) - Berlin and Frankfurt
- 2017-11-27: Corrected leap seconds in NAV-POSLLH.csv and RXM-RAWX.csv (Berlin and Frankfurt)
- 2017-10-20: Corrected Height above mean sea level (column 7) in BESTPOS.csv (Berlin and Frankfurt)
- 2017-10-20: Corrected number format in NAV-POSLLH.csv (Frankfurt)
Citation
@Conference{Reisdorf2016,
author = {Pierre Reisdorf and Tim Pfeifer and Julia Bre{\ss}ler and Sven Bauer and Peter Weissig and Sven Lange and Gerd Wanielik and Peter Protzel},
title = {The Problem of Comparable GNSS Results – An Approach for a Uniform Dataset with Low-Cost and Reference Data},
booktitle = {The Fifth International Conference on Advances in Vehicular Systems, Technologies and Applications},
year = {2016},
editor = {Markus Ullmann and Khalil El-Khatib},
volume = {5},
pages = {8},
month = {nov},
note = {ISSN: 2327-2058},
}
Copyright/License
This work is licensed under a Creative Commons Attribution - NonCommercial - Share Alike 4.0 International license.
All datasets published within this website are copyright by the members of smartLoc and published under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 License. Therefore, you may not use these datasets for commercial purposes and if you alter, transform, or build upon this work, you may distribute the resulting work only under the same license.
Privacy
These datasets are made available for academic use only. If you find yourself or personal data in these datasets and feel unhappy about it, please contact one of our team. We will remove the corresponding data as soon as possible.
Sensor Setup
The figures show the measurement vehicle Carai1 from the Professorship of Communications Engineering, the roof of the vehicle with the antenna setup and dimensions of the Carai1 with the positions of the antennas.
![Click to enlarge](gnss_dataset/figures/carai1_2.jpg)
Enumaration of all sensors for these datasets:
- Measurement Vehicle: Carai1 (Paper)
- Reference Navigation System: NovAtel SPAN Differential with RTK and IMU
- Mass-market GNSS receiver: u-blox EVK-M8T
- Ego motion data (velocity and yaw-rate) from CAN
- Additional: uncalibrated camera data
- [1] Guppy Pro F-125C
- Name: GuppyFront
- Resolution: 1288x960 pixels
- Frequency: 30 Hz
- Color: yes
- Direction: Front
- [2] Guppy F-036C
- Name: GuppyRear
- Resolution: 748x476 pixels
- Frequency: 30 Hz
- Color: yes
- Direction: Rear
- [3] and [4] - Aglaia INKA-V4 Stereo-Camera (Left and Right)
- Name: [3] AgliaFrontLeft and [4] AglaiaFrontRight
- Resolution: 640x480 pixels
- Frequency: 30 Hz
- Color: black/white
- Direction: Front
- [5] Aglaia INKA-V4
- Name: AglaiaRear
- Resolution: 640x480 pixels
- Frequency: 30 Hz
- Color: black/white
- Direction: Rear
Datasets
Berlin Potsdamer Platz |
Berlin Gendarmenmarkt |
Frankfurt am Main Main Tower |
Frankfurt am Main Westend Tower |
|
---|---|---|---|---|
Overview | ~1600 m / ~49% NLOS![]() |
~4950 m / ~37% NLOS![]() |
~2925 m / ~46% NLOS![]() |
~2340 m / ~32% NLOS![]() |
Data inside | Raw data GNSS sensor Reference GNSS data |
|||
Data format | Explanation Data formats | |||
Satellite systems | GPS / Glonass | GPS / Glonass / Galileo | ||
Download | GNSS Dataset (1.9 MB) |
GNSS Dataset (5.9 MB) |
GNSS Dataset (5.8 MB) | GNSS Dataset (4.2 MB) |
Video | Front Camera View | Front Camera View | Front Camera View | Front Camera View |
3D Overview |
Data formats
Possible formats: BestPos (NovAtel), BestVel (NovAtel), Receiver Position Fix (ublox), Raw Data (ublox), Yaw-rate, Velocity, Camera data, SP3 Data
Best position from NovAtel (filename: BESTPOS.csv) - Details on page 125 SPAN® on OEM6® Firmware Reference Manual (Revision 7) from NovAtel
Number | Description (Name) | Information source Reference Receiver (NovAtel) |
Unit |
---|---|---|---|
1 | GPSWeek | NovAtel | weeks |
2 | GPSSecondsOfWeek | NovAtel | s |
3 | Solution status (sol stat) | NovAtel | |
4 | Position type (pos type) | NovAtel | |
5 | Latitude (lat) | NovAtel | deg |
6 | Longitude (lon) | NovAtel | deg |
7 | Height above mean sea level (hgt) | NovAtel | m |
8 | Undulation (undulation) | NovAtel | m |
9 | Datum ID number (datum id#) | NovAtel | |
10 | Latitude standard deviation (lat σ) | NovAtel | m |
11 | Longitude standard deviation (lon σ) | NovAtel | m |
12 | Height standard deviation (hgt σ) | NovAtel | m |
13 | Base station ID (stn id) | NovAtel | |
14 | Differential age in seconds (diff_age) | NovAtel | s |
15 | Solution age in seconds (sol_age) | NovAtel | s |
16 | Number of satellites tracked (#SVs) | NovAtel | |
17 | Number of satellites used in solution (#solnSVs) | NovAtel | |
18 | Number of satellites with L1/E1/B1 signals used in solution (#solnL1SVs) | NovAtel | |
19 | Number of satellites with multi-frequency signals used in solution (#solnMultiSVs) | NovAtel | |
20 | Extended solution status (ext sol stat) | NovAtel | |
21 | Galileo and BeiDou signals used mask (Galileo and BeiDou sig mask) | NovAtel | |
22 | GPS and GLONASS signals used mask (GPS and GLONASS sig mask) | NovAtel |
Best available velocity data from NovAtel (filename: BESTVEL.csv) - Details on page 395 in OEM6® Family Firmware Reference Manual (Revision 9) and page 116 SPAN® on OEM6® Firmware Reference Manual (Revision 7) from NovAtel
Number | Description (Name) | Information source Reference Receiver (NovAtel) |
Unit |
---|---|---|---|
1 | GPSWeek | NovAtel | weeks |
2 | GPSSecondsOfWeek | NovAtel | s |
3 | Solution status (sol status) | NovAtel | |
4 | Velocity type (vel type) | NovAtel | |
5 | A measure of the latency in the velocity time tag in seconds. It should be subtracted from the time to give improved results. (latency) | NovAtel | s |
6 | Differential age (age) | NovAtel | s |
7 | Horizontal speed over ground (hor spd) | NovAtel | m/s |
8 | Actual direction of motion over ground (track over ground) with respect to True North | NovAtel | deg |
9 | Vertical speed, where positive values indicate increasing altitude (up) and negative values indicate decreasing altitude (down) | NovAtel | m/s |
Receiver Position Fix (filename: NAV-POSLLH.csv) - Details on page 281 in u-blox 8 / u-blox M8 Receiver Description (Revision R11)
Number | Description (Name) | Information source (Ground truth (GT) / Mass-Market Receiver (ublox)) |
Unit |
---|---|---|---|
1 | GPSWeek | ublox | weeks |
2 | GPSSecondsOfWeek | ublox | s |
3 | Longitude (GT Lon) | GT | deg |
4 | Longitude Cov (GT Lon Cov) | GT | deg |
5 | Latitude (GT Lat) | GT | deg |
6 | Latitude Cov (GT Lat Cov) | GT | deg |
7 | Height above ellipsoid (GT Height) | GT | m |
8 | Height above ellipsoid Cov (GT Height Cov) | GT | m |
9 | Heading (0 = North) - (GT Heading) | GT | rad |
10 | Heading Cov (0 = East, counterclockwise) - (GT Heading Cov) | GT | rad |
11 | Acceleration (GT Acceleration) | GT | m/s^2 |
12 | Acceleration Cov (GT Acceleration Cov) | GT | m/s^2 |
13 | Velocity (GT Velocity) | GT | m/s |
14 | Velocity Cov (GT Velocity Cov) | GT | m/s |
15 | YawRate (GT Yawrate) | GT | rad/s |
16 | Yaw-Rate Cov (GT Yaw-rate Cov) | GT | rad/s |
17 | GPS time of week of the navigation epoch (iTOW) | ublox | ms |
18 | Longitude (lon) | ublox | deg |
19 | Latitude (lat) | ublox | deg |
20 | Height above ellipsoid (height) | ublox | m |
21 | Height above mean sea level (hMSL) | ublox | m |
22 | Horizontal accuracy estimate (hAcc) | ublox | m |
23 | Vertical accuracy estimate (vAcc) (-1 == No Information) |
ublox | m |
Raw Data (filename: RXM-RAWX.csv) - Details on page 309 in u-blox 8 / u-blox M8 Receiver Description (Revision R11)
Number | Name | Information source (Ground truth (GT) / Mass-Market Receiver (ublox) / Reference Receiver (NovAtel)) |
Unit |
---|---|---|---|
1 | GPSWeek | ublox | weeks |
2 | GPSSecondsOfWeek | ublox | s |
3 | Longitude (GT Lon) | GT | deg |
4 | Longitude Cov (GT Lon Cov) | GT | deg |
5 | Latitude (GT Lat) | GT | deg |
6 | Latitude Cov (GT Lat Cov) | GT | deg |
7 | Height above ellipsoid (GT Height) | GT | m |
8 | Height above ellipsoid Cov (GT Height Cov) | GT | m |
9 | Heading (0 = North) - (GT Heading) | GT | rad |
10 | Heading Cov (0 = East, counterclockwise) - (GT Heading Cov) | GT | rad |
11 | Acceleration (GT Acceleration) | GT | m/s^2 |
12 | Acceleration Cov (GT Acceleration Cov) | GT | m/s^2 |
13 | Velocity (GT Velocity) | GT | m/s |
14 | Velocity Cov (GT Velocity Cov) | GT | m/s |
15 | YawRate (GT Yawrate) | GT | rad/s |
16 | Yaw-Rate Cov (GT Yaw-rate Cov) | GT | rad/s |
17 | Measurement time of week (rcvTow) | ublox | s |
18 | GPS week number (week) | ublox | weeks |
19 | GPS leap seconds (leapS) | ublox | s |
20 | Number of measurements to follow (numMeas) | ublox | - |
21 | Receiver tracking status (recStat) | ublox | - |
22 | Pseudorange measurement (prMes) | ublox | m |
23 | Carrier phase measurement (cpMes) | ublox | cycles |
24 | Doppler measurement (doMes) | ublox | Hz |
25 | GNSS identifier (gnssId) | ublox | - |
26 | Satellite identifier (svId) | ublox | - |
27 | Frequency slot - only Glonass (freqId) | ublox | - |
28 | Carrier phase locktime counter (locktime) | ublox | ms |
29 | Carrier-to-noise density ratio (cno) | ublox | dBHz |
30 | Estimated pseudorange measurement standard deviation (prStdev) | ublox | m |
31 | Estimated carrier phase measurement standard deviation (cpStdev) | ublox | cycles |
32 | Estimated Doppler measurement standard deviation (doStdev) | ublox | Hz |
33 | Tracking status (trkStat) | ublox | - |
34 | NLOS (0 == no, 1 == yes, # == No Information) | NovAtel | - |
Yaw-rate (filename: YAWRATE.csv)
Number | Name | Information source (Ground truth (GT) / Odometry (CAN)) |
Unit |
---|---|---|---|
1 | GPSWeek | GT | weeks |
2 | GPSSecondsOfWeek | GT | s |
3 | Yaw-rate | CAN | rad/s |
Velocity (filename: VELOCITY.csv)
Number | Name | Information source (Ground truth (GT) / Odometry (CAN)) |
Unit |
---|---|---|---|
1 | GPSWeek | GT | weeks |
2 | GPSSecondsOfWeek | GT | s |
3 | Velocity | CAN | m/s |
Camera data (filename: [GuppyFront, GuppyRear, AglaiaFrontLeft, AglaiaFrontRight, AglaiaRear].csv)
Filename format of figures: GPSWeek_GPSSecondsOfWeek.png
Number | Name | Information source (PC System / Camera) |
Unit |
---|---|---|---|
1 | GPSWeek | PC System | weeks |
2 | GPSSecondsOfWeek | PC System | s |
3 | Figure number (ongoing) | Camera | |
4 | Filename of figure | PC System |
SP3 data (filename: gmbGPSweekGPSDayofWeek.sp3.Z)
More explanations to the file format will you find on website of IGS MGEX GFZ service.
Information to Ground truth data
We generated the ground truth by fusing the BESTPOS message of the NovAtel receiver - Details on page 385 in OEM6® Family Firmware Reference Manual (Revision 9) - with the ego motion data from the CAN sensor of our test vehicle Carai 1.