Relative Position of Two Ultimate GPS Modules

Adafruit Ethernet, Motor, Proto, Wave, Datalogger, GPS Shields - etc!

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
User avatar
kevinpeterson
 
Posts: 2
Joined: Fri Mar 25, 2022 4:57 pm

Relative Position of Two Ultimate GPS Modules

Post by kevinpeterson »

I'm looking to get an accurate distance/direction from one Ultimate GPS module to another. They will be less than 100m apart, and I will need <1m accuracy. I don't care about absolute position, all I need know that module 1 is X meters / Y degrees from module 2. The two modules will have radio communication between them, will be outdoors with a clear view of the sky using the built-in antenna.

I have the system set up but I'm still seeing errors >1m, sometimes much more so. I've noticed that even if the two modules are very close together, sometimes they use a different number of satellites. I'm wondering, is there a way to sync the two such that both GPSs are using the exact same satellites/config? Since they are so close, I would expect then the relative error to be very small because whatever error from the signal would be the same for both modules. But I assume this expectation would hold only if the two modules are using the same satellites, etc.

So two questions I guess: (1) am I wrong to assume <1m error is possible with the Ultimate GPS in this scenario, and (2) is there a way to sync the two GPS modules such that they are using the same satellites?

User avatar
sj_remington
 
Posts: 994
Joined: Mon Jul 27, 2020 4:51 pm

Re: Relative Position of Two Ultimate GPS Modules

Post by sj_remington »

(1) yes and (2) no.

The combined error in distance measurement between two independent position determinations is greater than the position error in either one. For the details, look up "propagation of measurement errors".

For accurate distance measurement between two GPS-identified locations, most people use an RTK GPS setup, like the C94-M8P or later. It uses a different technology to achieve +/- 10 cm position error, or better.

Overview of RTK GPS: https://learn.sparkfun.com/tutorials/wh ... ps-rtk/all

Comprehensive review of the C94-M8P here. https://www.calvert.ch/maurice/2017/08/ ... valuation/

User avatar
kevinpeterson
 
Posts: 2
Joined: Fri Mar 25, 2022 4:57 pm

Re: Relative Position of Two Ultimate GPS Modules

Post by kevinpeterson »

Interesting, thanks. I looked into RTK GPS but a bit too much $$ for my use case.

I guess I'm just confused as to why two GPS modules <100m apart wouldn't experience almost exactly the same errors (and thus be able to calculate their relative position to each other very accurately). Isn't that the concept behind DGPS? That atmospheric conditions/obstructions will be very similar if the GPSs are close? I'm not disagreeing with you, as my experiments align with your conclusions. Is it really because the two modules are using different satellites, or are there other sources of error?

User avatar
sj_remington
 
Posts: 994
Joined: Mon Jul 27, 2020 4:51 pm

Re: Relative Position of Two Ultimate GPS Modules

Post by sj_remington »

Two seemingly identical GPS modules nevertheless show many internal differences. There are internal sources of noise in the receiver circuitry, as well as small differences in antenna construction, orientation, receiver sensitivity, etc.

On top of that, the wavelength of satellite transmissions is very short. Atmospheric disturbances and interference patterns caused by nearby reflecting and absorbing objects can lead to large differences in signal strength, even for identical antennas placed just meters apart.

Locked
Please be positive and constructive with your questions and comments.

Return to “Arduino Shields from Adafruit”