Welcome, Commercial Drone Pilots!
Join our growing community today!
Sign up

Litchi Advice

I understand what you are saying and what you do works for you, but manually adjusting the camera pitch and yaw when you have 20 shots to complete as accurate as possible every time you hit the site is much easier with controlled headings and POI's in my experience. Of course it's not tripod accurate and not even RTK is going to be, but if the error is only in a sphere of about 3m and the rest controlled by a robot it is pretty darn accurate.
 
  • Like
Reactions: Avocet
My comment above about manual camera contorl was actually more aimed at capturing video than stills.
That makes more sense, but the same applies in our use. Static video path used to make a floating timelapse that goes back and forth in time. The frames need to be as seamless as possible.

What are you controlling manually? The drone flies an autonomous waypoint mission, either uses auto-yaw or has to be given a heading so you are controlling gimbal pitch? Taking pictures while videoing?
 
What are you controlling manually? The drone flies an autonomous waypoint mission, either uses auto-yaw or has to be given a heading so you are controlling gimbal pitch? Taking pictures while videoing?

Using POI's and letting Litchi point the camera works very well too. But, you can have Litchi fly the programmed waypoint mission and leave the camera control, via yaw and gimbal pitch, to the operator. Not really worth the effort for construction documentation in most cases, but for a cinematic clip it can make a difference of going from fair or good composition of the subject to good or very good composition.

When you use POIs, you can still input correction but the program will tend to fight you and you could end up with a worse result. Next time you have a little extra time, program an orbit but leave the camera control set to manual and let Litchi fly the orbit and you focus on keeping the composition set exactly how you like it. I think you'll enjoy it.
 
  • Like
Reactions: adm_geomatics
I'll definitely try that! When I need that much control I usually free-fly, but something in between the worlds seems like it would get rid of some of the little unrefined issues I experience on both ends with the more creative videos. Thanks!
 
  • Like
Reactions: Dave Pitman
Give it a try sometime. After a little practice, you can often get better composition manually than the app will provide. Another skill for the kit!
Although, note though that I am coming from the perspective that it is the mission PLANNING on the hub that is the real art. I have a high degree of confidence in the Litchi app and as long a that remains, the onus of subject capture, interest and artistic endeavor lies solely on my mission planning, my editing skills, my creative self and command of the technology.
 
Those that really enjoy autonomous mission planning as a tool or art really should check out Dronelink.

It really is more capable than Litchi, but requires a mindset reset with a steeper learning curve. It will expand your creative mission planning.
 
Although, note though that I am coming from the perspective that it is the mission PLANNING on the hub that is the real art. I have a high degree of confidence in the Litchi app and as long a that remains, the onus of subject capture, interest and artistic endeavor lies solely on my mission planning, my editing skills, my creative self and command of the technology.

It's all good. :)

Someday, if you have a project where you don't have the luxury to sit down and pre-plan something like when showing up on site and being directed by a DP or client in the moment. Being able to integrate autonomous control at all becomes a luxury. If that ever happens, then you might have the opportunity to give this a try.
 
Those that really enjoy autonomous mission planning as a tool or art really should check out Dronelink.

It really is more capable than Litchi, but requires a mindset reset with a steeper learning curve. It will expand your creative mission planning.
The only reason you can say that Dronelink is more capable is because it has a few automated planning modes like any other mapping software out there. Litchi is much more granular than Dronelink or other mapping softwares so better and/or capable could be subjective. I have planned every mission that they show on their site except for the spiral in Litchi. I guess it would be handy if I was trying to model a Christmas tree or give someone vertigo. Don't forget the disclaimer that it is at least $100/yr for commercial use which I would assume is the majority of people on this site per the intent.
 
The only reason you can say that Dronelink is more capable is because it has a few automated planning modes like any other mapping software out there. Litchi is much more granular than Dronelink or other mapping softwares so better and/or capable could be subjective. I have planned every mission that they show on their site except for the spiral in Litchi. I guess it would be handy if I was trying to model a Christmas tree or give someone vertigo. Don't forget the disclaimer that it is at least $100/yr for commercial use which I would assume is the majority of people on this site per the intent.
I participated in the dronelink beta and thought they had some good ideas, though the interface needed to mature (at least at the time I beta'd it). When they went live, I thought they'd invite me to a copy of the platform -- they didn't. They did ask me to post my results though - pro bono - ... ummm. Lastly, the one thing Litchi has over most of its competitors is its execution model (UAV RAM) and not downlink comm. I like that fact that I could pack up and go home but the mission executes (if it actually can).
 
  • Like
Reactions: adm_geomatics
The only reason you can say that Dronelink is more capable is because it has a few automated planning modes like any other mapping software out there. Litchi is much more granular than Dronelink or other mapping softwares so better and/or capable could be subjective. I have planned every mission that they show on their site except for the spiral in Litchi. I guess it would be handy if I was trying to model a Christmas tree or give someone vertigo. Don't forget the disclaimer that it is at least $100/yr for commercial use which I would assume is the majority of people on this site per the intent.

Don't know what you mean by more "granular." I guess it is "subjective" since over the last nine months I have seen Dronelink develop many capabilities that are not in Litchi. It does take time and effort to figure out what is there and how features can be combined and controlled to do those things.

Those that spend the time to really look I think will come to the same conclusion.

But those that find Litchi will do all they need or want have no reason to look.

Stay safe and happy flying,
 
Don't know what you mean by more "granular." I guess it is "subjective" since over the last nine months I have seen Dronelink develop many capabilities that are not in Litchi. It does take time and effort to figure out what is there and how features can be combined and controlled to do those things.

Those that spend the time to really look I think will come to the same conclusion.

But those that find Litchi will do all they need or want have no reason to look.

Stay safe and happy flying,
I have used Dronelink and find it over-complicated and not a fit for at least 50% of drone users in these forums. It is completely impractical for mapping and too cumbersome for the types of autonomous videography missions most people use here. Especially when the user wants to maintain a high level of control. They are coming from the development, custom coding actions and a mindset of teaching a drone to do anything and everything possible. Honestly allot more intense than other softwares which technically I guess make it more "capable", but there is a point where the software needs to efficiently do what it is told in a matter of minutes and not creating a bunch of components and actions. What I mean by granular is that if I want to create a video flight plan with 20 waypoints all with different headings, elevations of focal points I can do it in about 5-10 minutes. If I want actions on those they are drop-down menus. You can batch edit on the fly. If I want a 360 pano on the fly from any point on the site I make two taps and it does it. I tried the mapping flight plan in Dronelink several times for both facade and terrain. Not even close. If they want in this space they better start thinking about practicality and enterprise type workflows and not coding a drone to do 360 barrel rolls and spirals. This is the kind of stuff you learn at Code Ninjas...
 
I don't use Dronelink for all missions especially the simpler ones; so I can appreciate some of your commentary.

There are other mission types that are easily set up within minutes that have unique capabilities. Orbits {multiple or partial) and spirals with increasing or decreasing radius and elevation changes from start to finish along with a select-able POI location are just a few of those that have been added recently.
The ability to set some of these up On The Fly was also recently added.

The flexible commands can be saved and plugged into future missions with just a few clicks eliminating having to repeat the setup of useful sequences.

They are different tools in the tool box.

It is nice to have choices.
 
Last edited:
You have two waypoints set up with non curved and the remainder curved. Problem is I don't believe you can get it to perform other actions at the non curved waypoints because you have set the initial setting to curve, and just eliminated the curve at the way-point setting.

Was able to fly a new test mission today, and as you suggested had the curved setting off. It took photos at it's waypoints as expected. On the first one it wasn't pointed the direction it was supposed to be but I'll let that go for now, probably my error. I added a video trigger after it took 8 photos, and it started and stopped as expected.

I had pauses built in before the photo, hoping to be in a stable hover when the shutter fires. It may not matter a lot though I think it can fire on the go just as well. I guess it will depend on the importance of the photo vs. efficiency.

I think a video mission needs to be it's own, you cannot turn on curved for individual waypoints, looks like a global setting. I watched the video with the jerky corners, not good. But it's not a problem, if you need both it's easy enough to terminate one mission in a hover and load the next mid-flight.

Actually today I was tempting fate a bit, ran a Pix4D mapping mission, when it returned I paused it before it landed, killed Pix4D, loaded Litchi, and sent it off mid flight, no touchdown. The app swap took a second and I had to pull the USB and re-insert, then it picked up the feed and I sent it on it's merry way. Of course all this was about 5ft over the landing pad but it handled the transition well. Kinda a practice run for if I ever have a site that needs mapping and some general photos together.
 
Was able to fly a new test mission today, and as you suggested had the curved setting off. It took photos at it's waypoints as expected. On the first one it wasn't pointed the direction it was supposed to be but I'll let that go for now, probably my error. I added a video trigger after it took 8 photos, and it started and stopped as expected.

I had pauses built in before the photo, hoping to be in a stable hover when the shutter fires. It may not matter a lot though I think it can fire on the go just as well. I guess it will depend on the importance of the photo vs. efficiency.

I think a video mission needs to be it's own, you cannot turn on curved for individual waypoints, looks like a global setting. I watched the video with the jerky corners, not good. But it's not a problem, if you need both it's easy enough to terminate one mission in a hover and load the next mid-flight.

Actually today I was tempting fate a bit, ran a Pix4D mapping mission, when it returned I paused it before it landed, killed Pix4D, loaded Litchi, and sent it off mid flight, no touchdown. The app swap took a second and I had to pull the USB and re-insert, then it picked up the feed and I sent it on it's merry way. Of course all this was about 5ft over the landing pad but it handled the transition well. Kinda a practice run for if I ever have a site that needs mapping and some general photos together.

You are braver than I am. I would never shut down any app while the drone was still flying. I quite using Litchi to do panos because it just doesn't work well with the Inspire 2, at least not for me. Not only that I can get at least on more pano per battery swap by doing the panos manually.

I only shoot video or photos on a way point flight but not both. As for for the drone stopping at each way point, edit those little glitches out with a good video editor. On my videos I always use curved way points, just works better for me.
 
It sounds like there are different use-cases here that explains some of the different methods. All the methods have their scenario that they work well in. Our typical scenario for standard images and video are to fly for construction progress where the client desires the exact same shot every time over the course of 12-24 months on a biweekly or monthly basis. These images can easily be turned into a time-lapse that looks seamless so manual control is not an option. Some of the patterns can get pretty intricate and keeping your eye on the drone is the regulation you don't want to break. I can't fly in many of my environments looking at the screen that much.
 
I have used Litchi for the first time yesterday. I manually started recording before flight and received an aural and visual confirmation that the app was recording to the Mavic Pro 2 on my smart controller. I had auto record off and no instructions at each waypoint as I was using waypoints with curves so thought that would suffice....... No Files at all.
Tried using GO4 app to test capture card and all was perfect.
Contacted Litchi support but no response so far. Trawling through the manual but no explanation as yet.
Basically the drone flew a perfect profile using Litchi but did not record anything despite showing an active record icon on the controller, quite strange.
Regards Mick
Belfast N.Ireland
 
I have used Litchi for the first time yesterday. I manually started recording before flight and received an aural and visual confirmation that the app was recording to the Mavic Pro 2 on my smart controller. I had auto record off and no instructions at each waypoint as I was using waypoints with curves so thought that would suffice....... No Files at all.
Tried using GO4 app to test capture card and all was perfect.
Contacted Litchi support but no response so far. Trawling through the manual but no explanation as yet.
Basically the drone flew a perfect profile using Litchi but did not record anything despite showing an active record icon on the controller, quite strange.
Regards Mick
Belfast N.Ireland
Just in case you didn't go this route you can email [email protected].
 

Members online

No members online now.

Forum statistics

Threads
4,291
Messages
37,659
Members
5,991
Latest member
Boduku