Height telementry is off

[Sasquatch]

It will not cause a flyaway or other issue except possibly RTH altitude issues. Haven't thought about it much really.

The pressure sensor is part of the IMU which is inside the Naza-M V2.

That's currently $200+ USD.

Ok, thank you very much again!

Tell me, what is that little red light that is underneath the phantom, it's like a single LED light that occasionally flashes. Does that have anything to do with the pressure sensor? I apologize if any of my questions seem stupid, but, I am a new flyer and completely new to RC UAV as well. I did my research prior, but I haven't taken a 6 month class on all the ins and outs about these things either.

 

[N017RW]

FYI:

DJI's independent developed 6-DoF Inertial Measurement Unit(IMU), one of the most important components of Ace One, Ace Waypoint, WooKong-H/WooKong-H Lite, WooKong-M, WooKong-M Waypoint, Naza-M (DJI's autopilot system), can provide precise linear acceleration and angular rate value output.

Inertial Measurement Unit （IMU）contains a 3-axis accelerometer, a 3-axis angular velocity meter and a barometric altimeter.

Inertial Measurement Unit is used for measuring the object's three-axis attitude angle (or angular rate) and the acceleration of the device, and is mainly used for identify the flight attitude. The DJI IMU inner damping design makes the installation more simple.

Generally speaking the IMU needs to be installed on the center of gravity.

The IMU is usually used for motion control, such as on cars and robots. It is also used on the accasions which need the attitude for precision positioning, such as submarine, aircraft, missiles and spacecrafts' inertial navigation equipment, etc.

 

[N017RW]

Where does the barometric pressure sensor get its zero-reference? Anyone? Home point, but is that even defined before taking off?

Look back a few posts.

 

[N017RW]

It will not cause a flyaway or other issue except possibly RTH altitude issues. Haven't thought about it much really.

The pressure sensor is part of the IMU which is inside the Naza-M V2.

That's currently $200+ USD.

Ok, thank you very much again!

Tell me, what is that little red light that is underneath the phantom, it's like a single LED light that occasionally flashes. Does that have anything to do with the pressure sensor? I apologize if any of my questions seem stupid, but, I am a new flyer and completely new to RC UAV as well. I did my research prior, but I haven't taken a 6 month class on all the ins and outs about these things either.

No it is for binding the Tx to Rx.

 

[M.Tigelaar]

In the beginning, my height is pretty good (+\- 0,5 meter) but if you wait on the ground the altitude go's up (3 to 5 meters) :shock:
It seems to be a tempeture thing.
I read that somwhere else
It is beter after power off the bird and then on.

 

[8676309]

Look back a few posts.

Is this what you're referring to below?

It's a relative measurement compared to the atmospheric pressure prior to take-off.

OK, *when* exactly is this "prior to take-off" event measured?

It seems like the OP is seeing 15-20' inaccuracy in height, I'm guessing before even taking off (wasn't specified). If this is so, then the process you describe is self-referencing and doesn't make much sense.

 

[Sasquatch]

It seems like the OP is seeing 15-20' inaccuracy in height, I'm guessing before even taking off (wasn't specified). If this is so, then the process you describe is self-referencing and doesn't make much sense.

Yes, 8676309, this is what I am saying, even before take off and again after getting the bird up in the air; the app is always reporting that she's approximately 15-20 feet higher than she is at any given moment.

 

[8676309]

It seems like the OP is seeing 15-20' inaccuracy in height, I'm guessing before even taking off (wasn't specified). If this is so, then the process you describe is self-referencing and doesn't make much sense.

Yes, 8676309, this is what I am saying, even before take off and again after getting the bird up in the air; the app is always reporting that she's approximately 15-20 feet higher than she is at any given moment.

Could you try an experiment and while it's sitting (before taking off) with the altitude showing inaccurate 15-20' off, flip S2 up and down about 5 times to re-set the home position. See if you have a new altitude reading after this.

 

[Sasquatch]

Could you try an experiment and while it's sitting (before taking off) with the altitude showing inaccurate 15-20' off, flip S2 up and down about 5 times to re-set the home position. See if you have a new altitude reading after this.

I sure can. I will do this tomorrow hopefully and if weather and time permits. Your suggestion sounds reasonable but something I wouldn't have thought of. I will let you know ASAP.

THANK YOU, for this suggestion.

Others have also helped to at least alleviate my fears that this little 'issue' won't cause my bird to slam into the ground for no apparent reason. I do like to know the reasons for things and I DON'T want my bird smashed to bits either so this is why I started this topic and asked the question.

Thanks again 8676309 and everyone who's replied. I truly appreciate ALL the feedback and suggestions.

 

[cahutch]

Zero reference is home point. Whenever that is set, automatically at power up or when manually reset during flight.

Keep in mind the zero reference isn't used for landing. At least not for precisely knowing where the ground is.
The barometer isn't accurate enough to be sure of where the ground is and neither is GPS altitude.
GPS altitude accuracy is +/- 15 meters best case. A barometer will vary as the weather changes, temperature, wind, etc, so it can't be trusted not to change during flight.
It's accurate enough when you're measuring height in hundreds of feet, but not for tens of feet.

The auto landing software figures out where the ground is on the fly. As it descends slowly it has no idea when it will touch the ground. It's continuously integrating altimeter changes moment by moment with the accelerometer and the engine power required to maintain a certain vertical decent speed.
In other words it knows it's on the ground when the altitude stops changing not when it thinks it's back to zero. This is why it always bounces a few times at touchdown. It's making sure it's on the ground. Probably using the accelerometers to detect the touchdown.

 

[Sasquatch]

Thanks again guys; Cahutch and M. Tigelarr for your repsonses...

In the beginning, my height is pretty good (+\- 0,5 meter) but if you wait on the ground the altitude go's up (3 to 5 meters) :shock:
It seems to be a tempeture thing.

M. Tigelarr, this is what I was wondering, perhaps the variables in reported height from my phantom aren't really unusual. But again, there is very little information about this on the discussion boards and nothing in the manual about this.

cahutch, thanks as well for your detailed information about this. Just to be clear though, on my 6th flight (the one before todays), I was trying to land my bird manually and directly on the ground, like I always have. You mentioned "auto landing" sequences and how the software should behave. Just wanted to be clear that on this 6th flight, I was attempting to land my bird manually.

Here's the 60 second video clip from that landing. You will see it bounce erratically and violently when I was attempting to do a manual landing. After seeing this, I raised her up and then landed her a few feet away without incident. And like I've mentioned, I've never had a problem like this before and with manual landings.

[youtube]https://www.youtube.com/watch?v=eC17-_J1-Qo[/youtube]

I wonder if this was just an anomaly and not a concern. Again, I'm mostly wondering about the 15-20 foot differential in reported height by the DJI app.

Thanks everyone for the responses and I'm going to try 8676309's suggestion and manually reset my home position while on the ground before my next take off and see if that helps.

BTW, I never expected that the telemetry would be bang on with a RC UAV, but regarding height stats, I assumed that these were a little more important for a RC helicopter.

 

[jspad]

I have a P1 (upgraded to 1.5 with iosd telemetry) and a P2V+ and I have never had better than intermittent accuracy for the height. Sometimes it is better than others. I believe it is a Naza issue. Just a guess.

 

[JWarren]

I notice some days I'm about 2ft off. Today I was right at 0 alt on take off though. Seems it could be an issue with a Fail Safe landing I would think?

Regards,
J.

 

[D_Tshudy]

Good thread. I actually learned something today!

 

[Mori55]

Zero reference is home point. Whenever that is set, automatically at power up or when manually reset during flight.

Keep in mind the zero reference isn't used for landing. At least not for precisely knowing where the ground is.
The barometer isn't accurate enough to be sure of where the ground is and neither is GPS altitude.
GPS altitude accuracy is +/- 15 meters best case. A barometer will vary as the weather changes, temperature, wind, etc, so it can't be trusted not to change during flight.
It's accurate enough when you're measuring height in hundreds of feet, but not for tens of feet.

The auto landing software figures out where the ground is on the fly. As it descends slowly it has no idea when it will touch the ground. It's continuously integrating altimeter changes moment by moment with the accelerometer and the engine power required to maintain a certain vertical decent speed.
In other words it knows it's on the ground when the altitude stops changing not when it thinks it's back to zero. This is why it always bounces a few times at touchdown. It's making sure it's on the ground. Probably using the accelerometers to detect the touchdown.

I LwYs thought it bounced because of ground effect with the prop wash.
My heights always alittle off too. Don't they have a magnetmometer that tells it when it's close to the ground ? I think I spelled it right.

 

[slothead]

Isn't the home point set when you take off? Because if so, then when you first start up the Phantom at a new location and connect (before taking off) the home point would be either undefined or the home point from the previous flight. Both of these references would be invalid.

That's correct, but that altitude is determined by the altimeter. The readout on the phone has to be calculated by subtracting the altitude of the home position from the altitude of the bird.

 

[jimdenburg]

I noticed the altitude discrepancy several months after I got my Phantom. It's normal for it to be off by some. Mine showed 20 ft. just before I was gonna hand land it and no I'm not that tall nor do I have the reach! From what I've read altitude isn't an exact science even on full scale aircraft. Taught me that I shouldn't go trying a fast pass @ 10 feet over the ice unless I have LOS. :shock:

 

[slothead]

Someone may have said that this was a good thread (and I agree).

I will add a tidbit (or two) in hopes of getting some valuable responses.

I haven't checked my height data since I started reading this thread, but I think mine is low at altitude. In analog functions there are two types of error: (1) offset error, and (2) gain error. With offset error the error stays constant throughout the range (e.g., of the Phantom showed it was off my 20' at launch, and stayed off by 20 ' at 100-200 feet or more). Gain error changes as the value changes (either negative or positive). Most analog error is a combination of both, a little offset and a little gain. So what I am hoping to hear from you guys that have reported errors, was it mostly gain error or mostly offset error? I'm guessing that this will be difficult to determine since at 200' it's pretty hard to tell if you are actually at 180' or 220'. But if you are reading 20' height when you are hovering 10' above the ground assuming no offset error (gain error = 2) and it's all gain error, at 200' you will be showing 400' height on the phone, that should be pretty easy to determine (maybe).

Anyone care to offer any feedback based on what they have experienced?

 

[M.Tigelaar]

Height increases as time passes even though he is all the time on the ground.
I had this with the Vision (not plus) and my second Vision plus.
two of the three :roll:
With me it is offset error I think.

 

[M.Tigelaar]

I found this even:

In my case, I only needed the thing to climb about 4 ft. before flying to the first waypoint, just enough to clear the deck rail. The instructions indicate it will rise vertically 16 ft. after take-off, THEN fly to the next waypoint. I haven't had a chance to try it again and that may well be what it does. In my first try it didn't but I'm not convinced it wasn't related to a different issue like the poor job I did calibrating the compass.

The barometric sensor in the Phantom sees rising temperature as an increase in altitude. There is a lot of heat generated inside the body just sitting there.

Mine always thinks it's 10 feet up after 3 minutes on the ground the first time it's powered on. If you've been messing around setting waypoints for much longer than that I wouldn't be surprised if the bird thinks it's already at 15 feet when you hit "Go". Just watch your camera screen for a few minutes on the ground at first power-on to see what I'm talking about.

it remains difficult :|