Awesome

u-blox-capture

Tools to capture RTK GPS output using u-blox C099-F9P application board.

Building

str2str is used to send RTK ground station information to u-blox device.

cd RTKLIB/app/str2str/gcc/
make

Usage

Configuration

ubx_configurator.py can be used to configure the u-blox device. example/ folder contains some example configurations that can be used. definitions/zed-fp9-interface-description.txt contains supported configuration values. It's not comprehensive, you can add missing values from ZED-F9P Interface Description. Use -flash flag store changes to on-board flash memory, making them persistent.

python ubx_configurator /dev/cu.usbmodem14123301 example/high_precision_gps_only.cfg

Setting up RTK

Connect u-blox device and find what device it is. On Mac for example:

ls /dev | grep cu.usbmodem

Next start str2str that connects to RTK ground station and delivers information to u-blox device.

You need to replace following with correct variables:

RTK information from your provider: USER, PASS, IP, PORT, MOUNTPOINT
LAT/LON: Rought estimate of current position
DEVICE: device name, for example: cu.usbmodem14123301

./RTKLIB/app/str2str/gcc/str2str -in ntrip://USER:PASS@IP:PORT/MOUNTPOINT -p LAT LON 0.0 -n 250 -out serial://DEVICE:460800:8:n:1

Leave this running on the backround.

u-blox device should now show a blinking blue light (connected to GPS) and solid yellow light next to it (connected to RTK).

Logging

Replace /dev/cu.usbmodem14123301 with your device.

python ubx_logger.py -v /dev/cu.usbmodem14123301

Converting to simple GPS coordinates

gps_converter.py is used to convert different UBX messages (PVT, TIMEUTC, HPPOSLLH) to accurate easy to use GPS coordinates. It takes the ubx-*.jsonl file created with ubx_logger.py as an input.

python gps_converter.py output/ubx-2021-01-16-17-22-16.jsonl

An example entry:

{
    "time": 1610810565.79965,    # UTC time in seconds
    "lat": 60.173692026,         # Latitude
    "lon": 24.803227458,         # Longtiude
    "altitude": 15.3108,         # Altitude in meters
    "accuracy": 10.364,          # Accuracy in meters
    "verticalAccuracy": 14.836   # Vertical accuracy in meters
    "velocity": {                # Velocity estimate in NED coordsys
        "north": -0.028,         # Meters
        "east": -0.056,          # Meters
        "down": -0.024           # Meters
    },
    "groundSpeed": 0.063,        # 2D ground speed in meters
    "speedAccuracy": 0.108       # Speed estimate accuracy in meters
}