https://github.com/alancoding/gpxstats
Project to gobble up some personal Garmin navigator data
https://github.com/alancoding/gpxstats
gps-unit gpx
Last synced: about 1 year ago
JSON representation
Project to gobble up some personal Garmin navigator data
- Host: GitHub
- URL: https://github.com/alancoding/gpxstats
- Owner: AlanCoding
- Created: 2016-02-12T01:03:29.000Z (about 10 years ago)
- Default Branch: master
- Last Pushed: 2017-01-29T21:31:09.000Z (about 9 years ago)
- Last Synced: 2025-01-03T05:39:15.678Z (over 1 year ago)
- Topics: gps-unit, gpx
- Language: Python
- Size: 37.1 KB
- Stars: 0
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
[](https://travis-ci.org/AlanCoding/gpxstats)
# GPX Stats Project
This is my project to gobble up some personal Garmin navigator data.
Its point is to be a data grinder for anyone who owns a navigation unit
and drives around with it regularly. The kinds of questions it hopes to
eventually answer are things like:
- Where places did you visit most?
- What speeds do you typically drive at?
- What are the highest elevations you drive to?
- What's the hardest you've ever braked?
### Site Map
For research about some related resources, see [the notes page](docs/Notes.md).
For some pretty graphs see [the data analysis examples](docs/data_examples.md).
## Instructions
Did you know that your GPS unit is probably storing a record of everywhere
you go? Creeped out? Turn those lemons into lemonade today.
### "In Real Life" Instructions
Plug in your Garmin unit to your computer and copy over the files in the
gpx archive folder into the archive folder of this project. These should
follow a filename pattern like:
- 1.gpx
- 2.gpx
- 3.gpx
The unit will store these sequentially as you drive, but it will delete
old ones as it accumulates new data. Thus, I can't predict what numbers
you will see.
### Python Instructions
You must put them in the archive folder with this naming. Then you can run the
analyzer with `python analyze.py`. Or, you can import the python module, and
then use it kind of like this:
```python
from gpxstats.objects import Archive
archive = Archive('archive/')
for pt in archive:
print(pt.speed)
```
#### Using the generator
The above example is the class-based implementation, and in addition to that
there is a generator-based implementation in order to provide a lighter
option to cycle through points.
## Questions
How can we take our extremely obvious intuition and apply it to a large set
of data like this? Some boundaries are easy to draw. I'm pretty sure my car
never went 1,000 miles per hour, so if we see that, it's safe to say that we
can mark some points as invalid. But our intuition is much more detailed than
this.
## Discoveries about Archive Character
Understanding exactly what type of thing a GPS unit stores is a trial and
error process.
### Packaged speed values
The Garmin units appear to record speed only in specific increments. These
increments seem to be 1.37 or 1.38, depending on how the rounding works out.
Here are a few of the first set of speeds that the unit uses:
1.37, 2.75, 4.12, 5.49, 6.86, 8.24, 9.61, 10.98, 12.35, 13.73, 15.1, 16.47
In terms of units, there are almost certainly in terms of meters per second.
### Data Imperfections
As of version 0.1, we can get a list of the top attributes for various fields.
Many of these just demonstrate where the data is bad. I will explain this
point-by-point.
#### Elevation
I found about 9 clusters of highly negative elevation values. Again, no,
I did not drive into Death Vally or something real like that. These are data
artifacts that will eventually need to get ironed out somehow.
## Roadmap
Future features:
- Data cleaning process that will drop a point if there is something
suspicious about it
- Segment-based objects that will store information about driving between
stops
- Trip and destination based objects
The final point is the spiffy ultimate goal, but the intermediate steps
remain a little more murky. Should we store all segments in an array for
the entire archive? That's not entirely clear.