GPS HDOP Range

Why does APM 2 complain about a HDOP value below 100 - this makes absolutely no sense since a position fix with a HDOP of 100 would be next to useless - lower = better in HDOP terms

What am I missing ?? is the Pixhawk some how scaling this value ??

100 to 900 for a HDOP range just makes no sense, as I understand HDOP

1= Ideal
1 to 2 = Excellent
2 to 5 = Good
5 to 10 = Marginal
10 to 20 = Poor
20 and above = May as well be on Mars - you have no clue where you are

It’s not AP2, but the autopilot sending the message to us. AP2 is pretty stateless on most things, it’s more a reporting tool.

There’s an HDOP check to be less than 2m (Horizontal Dilution of Precision). This helps make sure the GPS has at it’s best guess an accurate lock. That said, it’s an indicator, and is not an exact measurement. You could fly with a HDOP > 2m and everything works out just great. What it does indicate is that there is more potential for a GPS glitch. This can result in a dramatic change in where the copter thinks it is, and then a quick dash to correct that. This can result in a crash etc (object in the path to the new location). APM has glitch detection to avoid the worse case scenarios. This is one advantage on planes and rovers, they can just move randomly in any direction so momentary glitches are easy to ignore.

Appreciated - but the minimum setting on the APM parameter for what is an ‘acceptable’ HDOP is 100 … try putting anything else in like 2 or 4 etc and APM 2 bitches at you - in fact it doesn’t seem to save the parameter unless it is within its 100 to 900 range.

I’m trying to relate that to what I understand - because a HDOP of 100 is useless - you wouldn’t have any idea where you were.

does 100 = 1, 200 = 2 and so on ?? - because then it would make sense.

That is correct, per copter.ardupilot.com/wiki/common … S_glitches , a HDOP of 150 is 1.5, and 2.0 is 200, so the units are cm, rather than meters.

Perfick - this really is a learning cliff - not a curve.

I think for me the hard part is tracking stuff down and looking in the relevant places, I’d never have thought to look there - I was looking for something like a parameter WIKI - but the ‘comment’ on the parameter in the APM window doesn’t indicate ‘scale’ … just a comment from a noob … sorry if I’m being dumb.

@malcom2073 Are you sure the HDOP values are in units of distance and not pure numbers? I can’t find anything on the page you linked to indicate that they would be.

correct! HDOP has nothing to do with meters, feet or any linear measurement…
For a linear measurement look at Hacc

image
If HDOP, VDOP or PDOP (a combined factor) is high it indicates triangulation is poor. It is not a distance.

Thank you! So what exactly is Hacc?

Horizontal accuracy is the meaning

@callum1 Nope, I was wrong 5 years ago :slight_smile: it is not a distance, see the link below.

That being said, I’ve no idea how this applies to the value returned by APM anymore, it has been too long for me.

DOP is used as a mask for potentially bad GPS data.
If VDOP + HDOP = PDOP (I think this is right) gets over 5-6 you should not trust the data.
HDOP is not a good number to rely on alone but it will provide some accuracy assurance.

PDOP Definition

A-Horizontal Precision (m) B-Maximum PDOP Value. The Position Dilution of Precision ( PDOP ) is likely one of the most critical GPS AutoTrac values to monitor. As the PDOP value increases, both the horizontal and vertical precision (guidance accuracy) of your data points decreases

image
Above is a 3D fix with the M8P & 15 deg cutoff angle
image
Above is the M8P RTK float with 15 deg cutoff angle

No, For Hacc & Vacc they imply meters. For HDOP & VDOP the M is a possibly just code mistake

I think is it very rude to ask a question and delete it after you get an answer.

My apologies. Thank you for your patience and expertise. I was thinking of deleting my account to keep a cleaner trace of my internet profiles, but decided against it.

Thank you. I did not need to say anything no worries.