Litchi - Ground Relative Flights

[a31burgervan]

Has anybody used this yet and if so what are your thoughts before I take the plunge - hopefully not literally !


Sent from my iPhone using PhantomPilots

 

[smiller]

It works. You have to remember that all altitudes are referenced to your takeoff location, and you have to account for things not in the Google terrain database (trees, manmade structures, etc.) manually of course.

 

[goblues38us]

and trust google is correct

 

[smiller]

I think it should be plenty good enough for general flying, assuming you add some kind of a fudge factor as you always should. Probably some caution would be advised In areas with very rapid/steep terrain changes, and I sure wouldn't trust it for nap of the earth flying.

 

[a31burgervan]

Thanks fellas I think your thoughts mirror mine - the other thing a thought was to ensure plenty of waypoints in the mission to ensure height data was a accurate as possible with terrain


Sent from my iPhone using PhantomPilots

 

[dwallersv]

Thanks fellas I think your thoughts mirror mine - the other thing a thought was to ensure plenty of waypoints in the mission to ensure height data was a accurate as possible with terrain

It's a fantastic feature, and I've used it a bunch already. However, the cautions articulated in this thread are spot-on.

If you're planning to fly relatively close to the ground (100-150 feet for this feature) it's still very useful to export the path, take it in to Google Earth on a PC, and look at it in 3D to get a rough idea of obstacle clearance.

Buildings and other structures are accurate enough with Google Earth for elevation flight planning. Trees are not as accurate, but still much MORE accurate in terms of clearance height than the basic topological data Litchi is using.

So, if you need to fly relatively close to obstacles and follow varying terrain to maintain a constant AGL, Google Earth is still a very useful, and necessary tool. However, the mission planning can start in Litchi now, and it much easier, than the older Google Earth methodology.

Maybe we can get Litchi to make a GE map/planning feature available in-app? I don't know what the API looks like for GE -- there may not be a good way to do this, like there is with Google Maps.

 

[smiller]

No automated method is going to replace the human brain, at least not until the Phantom 12

Being able to plan a mission in advance is great, but no matter what method you use you still need to apply some manual oversight. Anyone who plans a mission in advance and just flies the bird upon arrival with no on-site review is going to eventually be without a drone.

 

[dwallersv]

Being able to plan a mission in advance is great, but no matter what method you use you still need to apply some manual oversight. Anyone who plans a mission in advance and just flies the bird upon arrival with no on-site review is going to eventually be without a drone.

Spot on.

I'm pretty new to this, but I follow a standard practice of flying missions first with heading set to Manual and gimbal pitch set to Disabled so I can look around while it flies and deal with any obstacles -- like treetops -- that couldn't be accounted for in the planning.

I've been too chicken to fly anything lower than about 150' AGL so I haven't actually had anything get in the way yet that I had to modify the mission for. But that's coming as I get more brave...

 

[BryanBTX]

No automated method is going to replace the human brain, at least not until the Phantom 12

Being able to plan a mission in advance is great, but no matter what method you use you still need to apply some manual oversight. Anyone who plans a mission in advance and just flies the bird upon arrival with no on-site review is going to eventually be without a drone.

Very true.

I plan ahead and on site I fly a few locations to make sure I'm high enough and nothing is in my way. Then fly the mission.

 

[smiller]

I plan ahead and on site I fly a few locations to make sure I'm high enough and nothing is in my way. Then fly the mission.

Yes, I usually perform a 'scout flight' as well, unless I'm over open desert or I am certain that I will be well above any potential obstacles. But if there's even the slightest doubt then I pre-run the course. Trust, but verify.

 

[bsartist]

I think it should be plenty good enough for general flying, assuming you add some kind of a fudge factor as you always should. Probably some caution would be advised In areas with very rapid/steep terrain changes, and I sure wouldn't trust it for nap of the earth flying.

When dealing with rapid/steep terrain changes, be sure to add plenty of waypoints because height is calculated at each waypoint and straight line ascents/descents are flown in between. If you have a long distance between waypoints and there is, say, a hump in between, you could hit it.