0

I'm working on a project which uses GPS coordinates. I have a UAV that flies, for example, over a meridian or a parallel. I suppose that the coordinates I'm reading from UAV are always increasing or always decreasing, but it isn't like that. In details, I can have for example 3 decreasing coordinates and 1 increasing coordinate, then (always for example) 10 decreasing coordinate and 1 or 2 increasing coordinates. Is that normal? It could be the precision of the sensor?

  • 1
    That will depend on the direction your UAV is flying, if it is going Straight North, no deviation from the North (0 Azimuth), then your longitude will remain constant only your latitude will change. I assume by GPS coordinates you mean latitude and longitude here, or are you working with northing and easting? – Hasan Mustafa Feb 28 '16 at 11:40
  • Yes I thought that one remains constant and the other increase or decrease, but sometimes the value that changes (lat or long) is like its "jump" back or forward. For example, suppose I have increasing coordinates like (10,15) (15,15) (20,15) (25,15) (of course these are fake coordinates for simplifying the example) and sometime I get (10,15) (15,15) (20,15) (15,15) (25,15). Is this normal? – bjorn Feb 28 '16 at 12:17
  • 1
    I would also say it depends on what is the projection you are using and what kind of increase you have ! can you give us those details ? Also have you tried to map them ? It would be interresting to see if it follows and general pattern or if it's definitly not realistic. – SIGIS Feb 28 '16 at 15:57
  • 2
    Your assumption is correct. You must have some noise in GPS measurement data which is normal especially if you have fix to only few satellites. – user30184 Feb 28 '16 at 16:45
  • 1
    @user30184 since my sensor take 10 to 100 coordinates at second, could be the average of these coordinates a good approximation in order to avoid the wrong value taken? – bjorn Feb 28 '16 at 17:20
1

There should be no reason for the coordinates to always increase or decrease. It would depend on the direction of the motion, if the UAV is traveling along a straight line you would expect the changes in coordinates to remain similar.

Keep in mind that GPS position estimates have high errors. A simple receiver using a pseudorange observable could have errors even as large as 20-30 meters depending on how it handles ionospheric and tropospheric delays. These errors will result in noisy position estimates, possibly causing the problem you are referring to.

I would suggest you look into using Kalman filters for smoothing your data. Here is a good interactive tutorial that you can use to get started with Kalman filters.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.