paparazzi-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Paparazzi-devel] GPS precision


From: address@hidden
Subject: Re: [Paparazzi-devel] GPS precision
Date: Tue, 4 Sep 2007 17:19:28 GMT

It's important to understand that DGPS merely improves the *accuracy* of the position estimate by subtracting any static error.  The only way to improve the *precision* of the GPS is by improving the antenna or the GPS module itself.  See http://en.wikipedia.org/wiki/Accuracy_and_precision for a detailed explanation of these terms.

It's also worth noting that Paparazzi uses a sort of DGPS already - by default!  With the standard "versatile.xml" or any other relative-referenced flight plan, all waypoints are specified in terms of X-Y-Z meters from the boot location.  So if you boot the plane in the same spot each time, all of your waypoints will be in exactly the same spot regardless of ionospheric conditions - at least until the next solar flare or ion cloud induces a shift.  This is exactly what DGPS does, except true DGPS would update the error correction much more frequently than once per flight.  If you want to use a true DGPS, a few lines of code would allow you to make the "Home" waypoint move dynamically with a connected ground GPS RX.  This would be useful for many purposes, but as I explained above, would not address precision and would not even have any significant effect on accuracy because the weather-induced shifts are very gradual and rare.

Finally, to answer your question :)  I have run multiple parallel ublox receivers for days at a time.  When configured for "car" or similar modes they do not drift more than a couple of meters but configured for 4g 3-D acceleration they do drift around 5m or so laterally, randomly and independently.  Every few hours there is a solar flare or something that slowly sends both receivers off 10-15m in some direction - in unison.  You should not be getting frequent 10-20m deviation unless you have a very small ground plane, RF interference, or multipath interference from a nearby buliding.  Also, for a heli, you should use "pedestrian" or "car" mode instead of the 4g airplane mode we usually use.


-- "Ethan Arnold" <address@hidden> wrote:
Dear List,

Has anyone ever tried laying 2 GPS receivers (same model) side by side and
comparing the output over a reasonable period of time?

I noticed (with one receiver, under good reception) the measured GPS
position slowly (under 1m/s) wanders around in an area of about 10-20m
diameter. I was wondering if a second GPS receiver of the same model would
show the same deviation at the same time. If this were the case, couldnt one
use the stationary receiver as a reference point to correct the mobile GPS,
as long as they are relatively close together (<1km), and hereby be able to
detect the slow horizontal drifting of a helicopter, which otherwise couldnt
be detected with only the on-board GPS receiver?

Regards,
Ethan

-----Urspr√ľngliche Nachricht-----
Von: address@hidden
[mailto:address@hidden Im Auftrag von
address@hidden
Gesendet: Dienstag, 4. September 2007 08:39
An: address@hidden
Betreff: Re: [Paparazzi-devel] Funjet

Hi Matt,

> Could you send a photo or explain how you mounted the brushless
outrunner to the
> FunJET if there are any special considerations?  You use the 2216/26
correct?

The Funjet comes with a plastic motor holder, the Axi is mounted to that
with screws, no magic here :-)

Make sure you carefully adjust the motor down thrust (in manual flight) as
described in the manual.

Martin





_______________________________________________
Paparazzi-devel mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/paparazzi-devel



_______________________________________________
Paparazzi-devel mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/paparazzi-devel


reply via email to

[Prev in Thread] Current Thread [Next in Thread]