Ya shoulda left out the drunk part ScottyT. REPORTED.Mate, with all due respect, I think you might be in the wrong forum, and drunk. Naza this, MC that.
I think the Phantom 3 might behave differently from what you are used to
Why is the drone home point when RTH never accurate
- Thread starter adjmirealty
- Start date
Oh please. I can't tell if you are joking or not but I think the latter.
The OP also said thanks to another's comment and you thought he meant you.
Try not to be so sensitive petal, it was a joke.
Ok. Just to clear things up.
On the Pilot app, left side, there is an auto take off (up arrow) which changes to a down arrow after take off. The Down arrow is now the auto land. If you hit this button, it will land in that spot.
Now in regards to RTH settings (triggered manually, lost of signal, low battery), there is 2 options (as stated before). You can have it ascend at a nominated height and then return home or you can simply just land at the spot you triggered RTH.
Meta4
Premium Pilot
- Joined
- Jan 8, 2014
- Messages
- 15,301
- Reaction score
- 9,243
Unfortunately consumer grade GPS does not give pinpoint accuracy.
Most of the time it is only accurate to within +/- 2 metres and sometimes more.
This can lead to a RTH being 2+2 metres out and is the major factor in the difference you are noticing.
While RTH and autoland is impressive, it's probably a good idea to resume control and land yourself. You can fly faster and choose a landing spot better than the P3 can.
Which leads to your second question...
Yes, you can resume control.
There are two full pages on RTH in the manual.
Reading the RTH section and practising is highly recommended.
- Joined
- Jul 10, 2015
- Messages
- 54
- Reaction score
- 3
- Age
- 54
What's a quad pad ?!?!My guess is that you haven't allowed sufficient time for the NAZA module to accurately record your home point. Do you give it sufficient time to record the home point?
Here's a quote from something which is not specifically for your aircraft but I think it's the same NAZA module:
"Home-point: Before takeoff, current position of multi-rotor will be saved as home-point by MC automatically when you start the motors for the first time after 6 or more GPS satellites are found (red light blinks once or no blinking) for 10 seconds."
From this:
http://download.dji-innovations.com/downloads/nazam-v2/en/NAZA-M-V2_Quick_Start_Guide_en.pdf
So, if you are not allowing enough time for the data to compute, like you get a go and liftoff right away, perhaps that's the cause.
You could test this by taking off as you normally do and heading out in one direction, then testing RTH to see if that's where it goes. To confirm, you could do the same but head out in a different direction and see if it thinks Home Point is out there.
Welcome to the party adjmirealty, hope you enjoy the show.
There are some helpful links in my signature.
Oh yea, add on a quadpad on your roof.
It's my idea of humor. A landing pad on the roof. You mentioned, I believe in jest, about your P3 landing on your house.
OK fine, so Scotty may be half right, I'm in the wrong forum, but if the OP got his questions answered; would someone like to tell me what he hell controls the P3 if it's not a NAZA module?
If the OP's questions aren't answered, please feel free to start a thread in OT titled "Educating the other guys".
If the OP's questions aren't answered, please feel free to start a thread in OT titled "Educating the other guys".
- Joined
- Jul 10, 2015
- Messages
- 54
- Reaction score
- 3
- Age
- 54
My roof is slanted ain't no quad pad for meIt's my idea of humor. A landing pad on the roof. You mentioned, I believe in jest, about your P3 landing on your house.
I think it's called "Pitched" and I know a guy... LOL.
Just kidding.
Did you get the answers to your questions? This can be a wonderfully informative place.
- Joined
- Jul 10, 2015
- Messages
- 54
- Reaction score
- 3
- Age
- 54
Thank you Ma'am, or Sir.
I have to agree whole heartedly. More fun than you can shake a stick at.
Ah-Ha!! So they slipped this paragraph in through a back door or something.
The Ver. 1.2 manual I've had stored on my phone and tablet since it was released doesn't say that. I thought you were making it up until I went to the P3 download page and looked at what's there. Same version as I've got but with a new paragraph.
I suppose I shouldn't complain now though since 2 months ago I complained on this forum that this "feature" was undocumented.
I love it when they change stuff like this on the fly and don't change the version. I wonder what else has changed in this "new" version of the manual.
OP when you say balcony what sort of position are you talking about ? side of a high building ?
try it in an open space and see how accurate it is without gps being blocked on one side
try it in an open space and see how accurate it is without gps being blocked on one side
ianwood
Taco Wrangler
10ft isn't good enough? Have you seen how big the planet is? Getting something to accurately return to within 10ft of it's origin is awesome!
Meta4
Premium Pilot
- Joined
- Jan 8, 2014
- Messages
- 15,301
- Reaction score
- 9,243
No .. compass and GPS are two very different things
The compass helps your Phantom fly straight and hold course.
GPS is what enables your Phantom to hold position and to find home.
GPS nard here.
GPS accuracy is not so good as many people think. If you have Android, install free "GPS status" then you can see the error probability of GPS. However good (maximum 8-10 GPS sats can be seen), it's around 2-3m; that is, the GPS coords differs 3m from real coords in 68% probability & 6m in 95% probability.
Also minimum resolution is few-10 feet. If Phantom stays rock solid (in few inches drift), it's by gyro/accelerometer, not by GPS. And GLONASS sats don't help to improve precision/resolution of GPS. They're independent systems.
GPS accuracy is not so good as many people think. If you have Android, install free "GPS status" then you can see the error probability of GPS. However good (maximum 8-10 GPS sats can be seen), it's around 2-3m; that is, the GPS coords differs 3m from real coords in 68% probability & 6m in 95% probability.
Also minimum resolution is few-10 feet. If Phantom stays rock solid (in few inches drift), it's by gyro/accelerometer, not by GPS. And GLONASS sats don't help to improve precision/resolution of GPS. They're independent systems.
Last edited:
IFly is correct, but the accuracy of unaugmented GPS is 16 ft (+/-). The more satellites you have, the more accurate it is. The OP's problem may be his home location- His balcony. The GPS signals may be getting bounced around by the house, the house may be shielding a couple of satellites from the Phantom. Just enough error introduced to displace the location by a meter or two.I'm not sure if I'm reading this correctly, but it's sounding like some folks say it should land "somewhere around" where it took off from.
Nope, it should land within a couple feet of where the "Home Point" is, and that's on a bad day.
On a good day, it should be within inches, not feet.
The accuracy you are seeing is totally normal for non assisted GPS. A 3-10 meter variation is not noticeable when navigating by GPS in a Car, Boat or Aircraft. But with our multicopters it is noticeable if the machine lands 2 meters from the take off position. It happens for various reasons, the computed position is always changing due to dynamic GPS satellite geometry. In the time between recording Home position and conducting the RTH you may have lost a satellite or two and picked up new ones as they rise and fall near the horizon. This will also cause tiny variations in computed position. Completely normal. Occasionally you might get a perfect RTH within inches but this will never happen every time.
I didn't read all the posts, but it seems like you have a bad calibration or something.
My RTHs always seem to be fairly close, and after reading this I went outside and taped down an X to launch from and tried various distances (50m away, 100m, 500m, etc) and they were all pretty spot on.
My RTHs always seem to be fairly close, and after reading this I went outside and taped down an X to launch from and tried various distances (50m away, 100m, 500m, etc) and they were all pretty spot on.
