RTH Phantom 3 Advanced - Error 3 mt. repetitive

Joined
Jun 18, 2019
Messages
7
Reaction score
0
Age
60
Location
ITALY
The hope is that, by sharing my puzzle with you, someone can help me ... In short:
I have a DJI Phantom 3 Advanced (last updated firmware, etc) and I really care about the effectiveness of RTH.
Well, despite the fact that he has repeatedly calibrated the compass, that the external conditions are ideal (only uninhabited plateau and very few metal parts in the vicinity, even 17 satellites are highlighted in the DJI GO status bar and repeatedly set to automatic, even manually, l 'Home Point on the red Landing Pad), inevitably, on automatic return, the drone always looks for the same point 2/3 meters to the left (west) from the pad where it started and where I repeatedly set the home point in the various tests . I read that when you have a connection with such a large number of satellites (17 in my case) the accuracy should be about 10 cm in terms of efficiency ... Here we are in the presence of an error that nature escapes me ...
Please give me some help ...
 
Joined
Jul 26, 2017
Messages
1,984
Reaction score
990
Location
Irvine (Orange County) CA
The hope is that, by sharing my puzzle with you, someone can help me ... In short:
I have a DJI Phantom 3 Advanced (last updated firmware, etc) and I really care about the effectiveness of RTH.
Well, despite the fact that he has repeatedly calibrated the compass, that the external conditions are ideal (only uninhabited plateau and very few metal parts in the vicinity, even 17 satellites are highlighted in the DJI GO status bar and repeatedly set to automatic, even manually, l 'Home Point on the red Landing Pad), inevitably, on automatic return, the drone always looks for the same point 2/3 meters to the left (west) from the pad where it started and where I repeatedly set the home point in the various tests . I read that when you have a connection with such a large number of satellites (17 in my case) the accuracy should be about 10 cm in terms of efficiency ... Here we are in the presence of an error that nature escapes me ...
Please give me some help ...
Great to test the RTH function. KInda like a fire drill.

Your P3A seems to be working just fine. The whole idea of RTH is to get the AC back to a point where you can re-take control and land the bird manually. I would love to read where it says the accuracy should be about 10 cm (4 inches).
 
Joined
Oct 14, 2016
Messages
3,853
Reaction score
1,698
Age
63
Location
UK Brit living Latvia
GPS positional accuracy is NOT cm's but m's ....

As to number sats received - NOT all are used to determine position - in fact quite a few are rejected when the GPS units calculate position.

If you take a position from a number of sats - you could be anything up to 10m or so in error. The usual expectation is 2 - 3m due to averaging.
If you were to have a plotter and leave it running in a static spot - the plot would amaze you - it starts to look like spots on a spiders web.
It gets even better !!

My yacht is 9knots max under engine ... about average 5knots under sail (its not a racer !). GPS plotter can have it at speeds in the 50 - 60knots and down to zero ... but because I set the averaging factor - I get reasonable speeds usually displayed with only occasional daft numbers. This is because of those positions giving speed / direction between.

As another says - RTH is a great tool to get the unit near home - for you the pilot to put her down accurately.

We do not have the military spec GPS (it works on different frequency) and even then it is not cms ... that accuracy comes about once optical guidance takes over from the GPS that has guided it to the location for example.
 
Joined
Jun 18, 2019
Messages
7
Reaction score
0
Age
60
Location
ITALY
It is ok ... but I still reflect on one thing ... If the average GPS error (with a high number of sat, obviously ...) is in the order of some meters and not cm ... we should have a random landing error ... In the sense that PH3 once lands on the right .. once on the left .. once forward ... etc etc .. of some meters.

In my case numerous tests have highlighted the fact that PH3 lands at about 2/3 meters always in the same point at a distance and in the direction of the red Pad Landig from where it started every time ...

I add a curiosity that perhaps is not of any importance: I heard from a video the experience of a PH3 pilot who had the same problem as me who said that he, setting the "Home Point" shortly after takeoff at 1.5 mt height ... and in this way you get the error of the next landing in RTH reduced to just a meter ...
 
Joined
Oct 14, 2016
Messages
3,853
Reaction score
1,698
Age
63
Location
UK Brit living Latvia
I agree that it is interesting that you are getting similar error and direction each test.

But I do advise that you do not let unit auto land - sooner or later you will end up tipped over or landing awkwardly. Much better to take over and manually land.
 
Joined
Oct 14, 2016
Messages
3,853
Reaction score
1,698
Age
63
Location
UK Brit living Latvia
IMU ?

Regardless to if it makes any difference to the RTH 3m you are getting ... my preferred way to calibrate :

For IMU - I regard same as compass etc. as the IMU has Compass in it ... see the bottom row of boxes in IMU display.

I prefer to do this in cold weather ... to reduce startup time later.
I find a nice clear flat level spot. I place a tile or wood plate making sure no nails or metal ... level it up with spirit level. Place my Phantom on the plate and double check its level.
Do the full IMU calibration ...

Once its completed - (some will argue the point here - but its my decision and I do it ... ) - I then do a Compass calibration to make sure that all interaction of Compass and IMU are at best possible.

Some people do IMU indoors ... that's their choice. I prefer outside in clear, calm area.

As they say - "Yer makes Yer choice !"
 

New Threads

Forum statistics

Threads
134,961
Messages
1,396,255
Members
98,347
Latest member
FelipeAlvarez