Litchi...

[bigloser2448]

Just dropping a note in this thread to say reading through this has convinced me to get the Litchi app. Thanks for sharing all your experience here. I've had my drone (P3S) for about 8 months or so. Been using DJI Go since I got it. I didn't know anything else existed until I found this forum. So far I've only been brave enough to try some POI flights relatively close, well within Signal strength. The Waypoint Mode doesn't seem very usable in DJI Go since you have to set each Waypoint while flying, or at least that's the way I think you have to do it. Setting up the Waypoint Mission ahead of time on the Computer sounds very exciting to me. I will experiment with some local waypoint missions well within my signal strength probably a dozen times before I get the nerve to go further. I only recently increased my maximum distance to 400M and have only flown to the limit once. (I'm a wimp).

Anyway, Thanks again. and to bsartist, I love your place up there in the mountains. What a paradise!!

 

[anotherlab]

I re-ran the mission at 11 miles per hour. Seems a bit better although it's not always reproducible. There were one or two noticeable gimbal movements but not the real jump like the last run. Also at the last waypoint it should have been pitched up to -1% to point an another POI a little off in the distance but it never quite pitched up that high. Don't know why. Here is the link. I caught a morning after a nice ice/snow fall

Did you lose connection at any point? That would affect the gimbal control.

 

[B Greene]

Does Litchi use Google maps? I really didn't want to have to install that "Here maps" on my tablet that the DJI Go app required.

Okay. I am still a newbie at with my P34K which my son and I got. He has flown a bunch and I have only got out once. We are flying with DJI Go and have not installed "Here maps". Not sure I would know how. What is it needed for, waypoint on DJI Go? Thanks.


Sent from my iPad using PhantomPilots

 

[bsartist]

Did you lose connection at any point? That would affect the gimbal control.

Nope.

 

[Brockrock]

I re-ran the mission at 11 miles per hour. Seems a bit better although it's not always reproducible. There were one or two noticeable gimbal movements but not the real jump like the last run. Also at the last waypoint it should have been pitched up to -1% to point an another POI a little off in the distance but it never quite pitched up that high. Don't know why. Here is the link. I caught a morning after a nice ice/snow fall

I think it looks awesome and has so much more impact a bit slower. I did see a few gimbal bumps which if you want perfection, you can edit out in post fading out and back in around them. This is a nice effect. Also, try 'Focus POI' on a mission. See what happens.

 

[bsartist]

I think it looks awesome and has so much more impact a bit slower. I did see a few gimbal bumps which if you want perfection, you can edit out in post fading out and back in around them. This is a nice effect. Also, try 'Focus POI' on a mission. See what happens.

Yeah. I like it slower (that's what she said ;-) ) I think my next step is to remove all extra waypoints along the straight lines and see if doing longer interpolation intervals might help. I'm leery about removing interpolation because I'd think it would be more jerky. IIRC your video didn't have many pitch changes so maybe the interpolation wasn't as important.

 

[anotherlab]

You could try Litchi support and send them the mission and a link to the video and ask them how to eliminate the glitches. If you could get that mission glitch free and 100% repeatable, you could do some really cool things with the video. Like run that mission every week for year and then fade in the videos is post production. Even if you only did August through October, you would get a great moving time lapse of the leaves changing color.

 

[bsartist]

You could try Litchi support and send them the mission and a link to the video and ask them how to eliminate the glitches. If you could get that mission glitch free and 100% repeatable, you could do some really cool things with the video. Like run that mission every week for year and then fade in the videos is post production. Even if you only did August through October, you would get a great moving time lapse of the leaves changing color.

That is EXACTLY what I'm trying to do. Very astute.

I was thinking more like once a month at most. To get some seasonal changes. But I think the chances of getting it precisely repeatable in terms of even lateral aim and levelness are slim. The answer would be to shoot in 2.7k and use the excess pixels to be able to crop and straighten down to 1080p so I can really register adjacent videos in the exact same spot and angle. Except my Mac won't handle 2.7k and I'm not dropping a few thou to upgrade that now.

 

[anotherlab]

That is EXACTLY what I'm trying to do. Very astute.

I was thinking more like once a month at most. To get some seasonal changes. But I think the chances of getting it precisely repeatable in terms of even lateral aim and levelness are slim. The answer would be to shoot in 2.7k and use the excess pixels to be able to crop and straighten down to 1080p so I can really register adjacent videos in the exact same spot and angle. Except my Mac won't handle 2.7k and I'm not dropping a few thou to upgrade that now.

You're right about the flight path never being 100% the same for each flight, but with the right transition effect it shouldn't be jarring to the user. That would let you stay at 1080P for recording. Which app are you using to edit the video with?

 

[bsartist]

You're right about the flight path never being 100% the same for each flight, but with the right transition effect it shouldn't be jarring to the user. That would let you stay at 1080P for recording. Which app are you using to edit the video with?

I use Final Cut Express on an early 2008 Mac Pro. Even with the 1080p everything has to be rendered which can be a pain. If I start doing any sizeable projects again I'd probably have to look at an upgrade. And either use Final Cut X or Davinci Resolve.

 

[JohnnyB]

I re-ran the mission at 11 miles per hour. Seems a bit better although it's not always reproducible. There were one or two noticeable gimbal movements but not the real jump like the last run. Also at the last waypoint it should have been pitched up to -1% to point an another POI a little off in the distance but it never quite pitched up that high. Don't know why. Here is the link. I caught a morning after a nice ice/snow fall

Beautiful place! Nice job!
Just a question: in the last POI you mentioned, did you set an altitude height in that POI's settings?

Sent from my XT1585 using PhantomPilots mobile app

 

[JohnnyB]

May not be relevant, but just may... I have always thought of every elevation setting in litchi of being related and based off of/to the original take off "0" altitude.. even POI altitudes. I believe you have to do the math just the same as waypoints. You'll have to look at the ASL of that POI in relation to the #1 waypoint ASL and do the math to get the desired POI altitude.
I'm saying this because I done a POI around an old barn that was lower at ground level than my takeoff point, and I had the elevation at the POI set a 5' but the camera pointed over top of the barn. That's when I discovered it was relevant to takeoff elevation.

Sent from my XT1585 using PhantomPilots mobile app

 

[bsartist]

Beautiful place! Nice job!
Just a question: in the last POI you mentioned, did you set an altitude height in that POI's settings?

Sent from my XT1585 using PhantomPilots mobile app

Thank you!

Yes I thought of the POI elevation. The ground level at POI 2 is, in fact, much lower than takeoff elevation. But I left the elevation at 3' above takeoff anyway because I wanted the camera pointed high in the air out to the lake. I double checked my assumptions by looking at the resulting gimbal angle itself at that waypoint. It was -1°. Which should be almost straight out. It's not close to straight out.

 

[JohnnyB]

Thank you!

Yes I thought of the POI elevation. The ground level at POI 2 is, in fact, much lower than takeoff elevation. But I left the elevation at 3' above takeoff anyway because I wanted the camera pointed high in the air out to the lake. I double checked my assumptions by looking at the resulting gimbal angle itself at that waypoint. It was -1°. Which should be almost straight out. It's not close to straight out.

Sounds like you're all over it. Like it was said earlier.. the tweaking of a mission is part of the fun! Your video already looks great.. can't wait to see the final.

Sent from my XT1585 using PhantomPilots mobile app

 

[bsartist]

Sounds like you're all over it. Like it was said earlier.. the tweaking of a mission is part of the fun! Your video already looks great.. can't wait to see the final.

Sent from my XT1585 using PhantomPilots mobile app

Thanks! ideally the final will be a year from now. When I stitch together footage from multiple runs throughout the seasons.

 

[JohnnyB]

Thanks! ideally the final will be a year from now. When I stitch together footage from multiple runs throughout the seasons.

That'll be a neat vid. I need to learn much more on the edit side. Just starting. Been mainly just flying low altitude/penetration flights just to be flying.

Sent from my XT1585 using PhantomPilots mobile app

 

[smiller]

May not be relevant, but just may... I have always thought of every elevation setting in litchi of being related and based off of/to the original take off "0" altitude.. even POI altitudes. I believe you have to do the math just the same as waypoints. You'll have to look at the ASL of that POI in relation to the #1 waypoint ASL and do the math to get the desired POI altitude.
I'm saying this because I done a POI around an old barn that was lower at ground level than my takeoff point, and I had the elevation at the POI set a 5' but the camera pointed over top of the barn. That's when I discovered it was relevant to takeoff elevation.

This is a good point to remember because it is non-intuitive. It took me a while to figure out by I was pointing above or below some POIs.

 

[anotherlab]

If you are going to do a Litchi mission over hilly terrain, you may want use Google Earth to create the mission. There's a great YT video that shows how to create mission where the waypoints altitude is relative to the location of the waypoint. You draw the path in Google Earth, save it as a KML file, and then load it into the Mission Hub to set the camera/gimbal settings.

 

[RodPad]

Wow, this thread is great, just found it.

Yes, the rising ground.
As some of you were part of this tread.
Litchi Mission Lost Phantom

Figuring out why he lost his Phantom, and finally understanding how @jiggyb21 was calculating where it could have disappeared, really took me out of the fog!

By taking his mission, AGL first way point, etc, doing the math on paper at every one of his way points, then adding way points, forced my to figure it out.

@JohnnyB
"This is why it's important to set your first waypoint relatively close to take off location"
Yes, I have been doing that, funny when you do the math it makes it easier.
"You'll have to look at the ASL of that POI in relation to the #1"
Yes, I hadn't got the figured out. Thanks again.

So this is scary one I had been plotting, -200 ft' -400 ft from first way point, to circle around friends house.
Mission Hub - Litchi
I'll try someday, but when I''m ready.
To confirm my subtracted calculations
I made a mission starting at his house to mine.


@Brockrock
"I would set the speed at 10 - 12 MPH"
Yeah, that's no brainier After somebody makes the statement.
"I think that acquiring this software any way other than downloading it from the proper source"
Yes, I think he got it loaded by some goofed link and should remove it and start over from google play.

I bought Litchi for Android and crApple on the same day, Android crashes when I open app, no big deal at the moment, I know I will use it on a newer Android in the future.

Everybody else, all good stuff!
I will pass out some likes.
And watch all of you guys, so if you followed by a dog, its not with my nose.

Rod

 

[JohnnyB]

So this is scary one I had been plotting, -200 ft' -400 ft from first way point, to circle around friends house.
Mission Hub - Litchi
I'll try someday, but when I''m ready.
To confirm my subtracted calculations
I made a mission starting at his house to mine.

Okay BOBO.. (laughs.. that's original.. like it!)
Your house (takeoff point) is 1069' ASL in your mission. ("Everything" is based off this altitude)
Your friends house, at the POI is 625' ASL. If you want the camera to point at his house, take the POI elevation up about 10' off the ground so it will point more towards the center of his house, (keeping it in the center of the screen) add 10' to his ASL gives you 635' ASL... So, 1069' - 635' = 434'. So -434' in your POI's altitude setting would point your camera at the center of his house.
The 80' you had set as POI altitude is over 500' above his house because its 80' above the 1069' ASL (takeoff point) making it 1149' ASL. His house is 625' ASL.
If you want your drone to be approx. 200' AGL at his house.... His house/625'ASL + 200' = 825' ASL ...... 1069 - 825 = 244... Waypoints around his house set at -244' you will be 200' AGL.

See how it works? Hope this helps.

EDIT: one more thing to check,.. between #1 & #2 waypoints you're dropping 300' in about a half mile.. remember it will start descending right from #1 so make sure it will clear those trees behind the house.. or at least keep a finger on the pause button.. just sayin. Probably not an issue.. but I would triple check.