to determine the altitude of the top of an obstacle

[engravosaurus]

Hi

If
1) I aim the camera directly forward and
2) fly to the top of an obstacle and
3) position the top of the obstacle in the center of the screen and
4) read the altitude

is that a accurate measure of the height of the obstacle or
is there distortion/whatever such that that is not an accurate measure

why do I care?
we have some tall fir trees on a hill and would like to set RTH altitude to be greater than the firs/hill
I could just say 200' and let that be that but
I flew to the top of the firs, put the leader in the center of the screen and read 106'
so am wondering how accurate is 106'?

thank you

 

[Jthomp]

I assume you are using GPS to obtain your altitude information, right? GPS altitude readouts were, at one time, purposely skewed so as to not allow the USA's enemies the opportunity to use the GPS system that the USA put into space against the USA in a war. That, however, was basically done away with when those same enemies put their own GPS system in orbit.

Nonetheless, the GPS altitude and geographical point locations are not always exactly dead-on for most applications, low-cost drones included (you can check this out by taking off in P-GPS mode, flying around for about 56 minutes, then having the drone do a RTH. It will climb to it's pre-programmed RTH altitude, after which it will proceed to a point above the take-off point (supposedly). It will then descend and eventually land, but it won't land precisely on the same spot that it took off from. It will be anywhere from just a few inches off all the way up to as much as 2 feet off (at least that has been my experience). [Just yesterday I took off from a side walk that was next to a flower bed. Five minutes later when I did a RTH, my P3S became an airborne weed-whacker-wanna-be when it got down to the last foot or so of flight!)

 

[Meta4]

is that a accurate measure of the height of the obstacle or
is there distortion/whatever such that that is not an accurate measure

It will be close to an accurate measurement of the height above the level where you launched but may be a few feet out.
But if you are using it for RTH calculations, you want to add a safety margin anyway so it should be good enough for that purpose.

 

[Meta4]

I assume you are using GPS to obtain your altitude information, right? GPS altitude readouts were, at one time, purposely skewed so as to not allow the USA's enemies the opportunity to use the GPS system that the USA put into space against the USA in a war. That, however, was basically done away with when those same enemies put their own GPS system in orbit.

That's a little confused.
There was an intentional error introduced to GPS (selective availability) but that was for horizontal position and it was switched off in 2000: GPS.gov: Selective Availability
GPS has always been (and still is) very inaccurate for altitude and for that reason, the Phantom doesn't use GPS data at all for altitude.
The Phantom uses a barometer to derive relative altitude from changes in air pressure.

 

[Alan-G]

I assume you are using GPS to obtain your altitude information, right? GPS altitude readouts were, at one time, purposely skewed so as to not allow the USA's enemies the opportunity to use the GPS system that the USA put into space against the USA in a war. That, however, was basically done away with when those same enemies put their own GPS system in orbit.

Nonetheless, the GPS altitude and geographical point locations are not always exactly dead-on for most applications, low-cost drones included (you can check this out by taking off in P-GPS mode, flying around for about 56 minutes, then having the drone do a RTH. It will climb to it's pre-programmed RTH altitude, after which it will proceed to a point above the take-off point (supposedly). It will then descend and eventually land, but it won't land precisely on the same spot that it took off from. It will be anywhere from just a few inches off all the way up to as much as 2 feet off (at least that has been my experience). [Just yesterday I took off from a side walk that was next to a flower bed. Five minutes later when I did a RTH, my P3S became an airborne weed-whacker-wanna-be when it got down to the last foot or so of flight!)

"56 minutes" where do you get your batteries from, I want some. Ha Ha

 

[With The Birds]

If you have the horizon in frame and it appears lower than the trees you are higher than them. Allow 10m safety margin for your RTH setting, better to be safe than sorry.

 

[wa5tef]

My Phantom 4's altitude reading is almost always off by several feet. I can fly a Litchi waypoint mission and at the end of the mission, the altitude readout and voice notification will give the altitude that is usually about 10 feet higher than it actually is, and upon landing, it will read about ten feet. I sometimes worry about the waypoint altitude I have programmed in on Litchi but so far, I have allowed about 15 feet and not had any crashes.

Thanks,
Jim
WA5TEF

 

[engravosaurus]

got it, thanks all

gets relative altitude from barometer
will repeat and see if horizon is in the frame (why? just for the fun of it)
will add ~10m for RTH (use 136' or round to 140')

 

[John Tarvin]

Went flying today and measured 6 trees using the "get close then center the screen on the highest point on the tree" method. Why? To plan a waypoint mission to fly "just" above the trees. We will be adding around 20 feet margin for measurement and altitude inaccuracies. If that's not enough, thank goodness for forward obstacle avoidance. Note: all trees are in full leaf, no twiggy issues. Will attempt the mission next week.

 

[engravosaurus]

Did you see the horizon comment in Post #6?
and
He recommended 10m or 30'.
and
What is the mission?

 

[John Tarvin]

Did you see the horizon comment in Post #6?
and
He recommended 10m or 30'.
and
What is the mission?

Yes, but kinda tough here as our horizon often gets interrupted by hills.

30' would be safe. Last Friday I used the "center on screen" method, added 8 to 20', put it in tripod mode (slow and all sensors on) and flew right over the top. Didn't trigger and obstacle warning.

The "mission" is to fly through and open field at about 100' and around 20mph using waypoints. As we approach a tree, descend in altitude to our flyover height, then fly over the tree at about 2-4mph. Ascend and repeat for next tree until all trees are done... Or we've crashed.

It'll for certain put a lump in our throat (or in our pants), but I've always liked dramatic shots flying over trees at tree top level.