- This project generates virtual fisheye zenith images from Google Earth at arbitrary locations and automatically determines GNSS visibility
- This can be used to evaluate GNSS multipath in urban areas and as a reference for LOS/NLOS satellites
- Virtual fisheye video generated from GNSS data mounted on the vehicle included in the dataset
- MATLAB (>R2022a)
- Image Processing Toolbox (for satellite_visibility_xxx.m)
- Google Earth Pro
- MatRTKLIB
- Install Google Earth Pro
- Clone or download MatRTKLIB and add to path in MATLAB
- Clone or download ge-gnss-visibility
ge_fov160.kml
is generated
- Check
Photorealistic
andTerrain
in Layers
- Select tour
fisheye_fov160
- Enter the name of the file to save to
- Video parameter:
Custom
- Picture size:
1200 x 1200
- Frames per second:
1.0
- Output configuration
- For kinematic data:
MJPEG (.mov)
- For static data:
JPEG image sequence (.jpg)
- For kinematic data:
- Usually this process takes quite a while!
- In my experience, running Google Earth in a Linux (Ubuntu) environment is much faster than running it in Windows
- On Windows, it is faster to use OpenGL instead of DirectX as the graphics mode for Google Earth
- The type of graphics card you have can also affect performance
- Virtual fisheye image/movie is generated
If you use this project in your research or paper, please cite the following article [link].
@inproceedings{suzuki2015simulation,
title={Simulation of GNSS satellite availability in urban environments using Google Earth},
author={Suzuki, Taro and Kubo, Nobuaki},
booktitle={Proceedings of the ION 2015 Pacific PNT Meeting},
pages={1069--1079},
year={2015}
}