https://github.com/filipspl/gps-accuracy
gps accuracy while running - shall we use foodpods?
https://github.com/filipspl/gps-accuracy
footpod forerunner garmin-watch gps gps-data
Last synced: 5 months ago
JSON representation
gps accuracy while running - shall we use foodpods?
- Host: GitHub
- URL: https://github.com/filipspl/gps-accuracy
- Owner: filipsPL
- Created: 2021-11-28T19:57:28.000Z (about 4 years ago)
- Default Branch: main
- Last Pushed: 2021-12-01T20:49:14.000Z (about 4 years ago)
- Last Synced: 2025-07-11T20:44:39.586Z (6 months ago)
- Topics: footpod, forerunner, garmin-watch, gps, gps-data
- Language: Jupyter Notebook
- Homepage:
- Size: 1.42 MB
- Stars: 3
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
GPS, footpod, and accuracy
======================
- [GPS alone vs GPS with footpod](#gps-alone-vs-gps-with-footpod)
- [Track](#track)
- [Hardware](#hardware)
- [Results: distance](#results-distance)
- [Results: speed variation during the run](#results-speed-variation-during-the-run)
# GPS alone vs GPS with footpod
**tl;dr: footpod increases the distance measurement accuracy of the gps sportwatch.**
## Track
Track distance: 1.71km +- 0.01km (3x measurements with garmin vista hcx and polar m 400 = 6 measurements), urban environment.
## Hardware
Hardware:
- Garmin Forerunner 945 (GPS+Glonass, 1s track saving)
- adidas Mi_coach footpod (ant+ version)
- Garmin Vista HCx (as a reference)
- Polar M400 (as a reference)
## Results: distance
tl;dr garmin with footpod measured almost the exact distance of the loop (1.71 km).
Three runs with three settings of garmin forerunner: (1) GPS+footpod, (2) GPS only, and (3) footpod only. All runs accompanied with two other gpses as a reference. Track distance: 1.71km +- 0.01km.
| Round | distance Polar (reference) | distance Vista (reference) | data source | distance | time | average pace | error | place |
| ----- | -------------------------- | -------------------------- | ----------- | -------- | ------ | ------------ | ------ | ----- |
| 1 | 1.72 | 1.70 | GPS+footpod | 1.71 | 9:40.5 | 05:40 | -0.10% | 1 🥇 |
| 2 | 1.70 | 1.71 | GPS | 1.67 | 9:31.8 | 05:43 | 2.24% | 3 |
| 3 | 1.72 | 1.70 | footpod | 1.68 | 9:28.7 | 05:38 | 1.66% | 2 |
- garmin with footpod measured almost the exact distance of the loop (1.71 km, -0.10%).
- garmin with gps only was the worst, with error +2.24%
- foot pod only (no GPS!) was 1.66% wrong
## Results: speed variation during the run
tl;dr: garmin+footpod offers more "stable" and consistent pace/speed
Raw data plots. Garmin device + polar m400 as a reference.
Smoothened data (5s window average). Garmin device + polar m400 as a reference.
Distribution of speeds:
And some statistics:
speed_calc
count
mean
std
min
25%
50%
75%
max
hardware
source
footpod
garmin
569.0
2.958910
0.105325
1.99
2.940002
2.960083
3.000000
3.639984
polar
574.0
2.999651
0.492067
0.00
2.900024
3.000000
3.199951
4.500000
gps
garmin
572.0
2.917657
0.770542
0.00
2.580013
2.894989
3.209970
7.880005
polar
576.0
2.957465
0.302476
0.00
2.800049
3.000000
3.099998
4.100006
gps+footpod
garmin
580.0
2.940086
0.085252
1.49
2.910004
2.949951
2.980042
3.070007
polar
606.0
2.852475
0.672028
0.00
2.799988
3.000000
3.100002
4.199997
**Results:**
- garmin with footpod: very stable pace.
- garmin with gps only: high variability of speed
- foot pod only (no GPS!): very stable pace.
- polar m400: pace is more stable than forerunner in gps-only mode. This may be due to averaging [citation needed].