I am a bit new to droning and using the Litchi app, which I find to be a great tool, and I have a question about Litchi. I have used the Litchi app on a couple of jobs, one a golf course and the other for real estate. I also teach an aviation class at a high school and had my kids use the app to do a mission around the school. What I have found is that the Litchi app does not do a very good job of keeping the point of interest centered in the frame. It almost always seems to be off to one side or the other so I wind up having to do several test flights and then adjusting the position of the point of interest to try to keep the camera focused. This was very apparent on the golf course job. I would fly my Phantom down the fairway with the green as my point of interest only to find that by the time the drone reached the green, it was off to one side rather than focused on the point of interest. My students had the drone fly around the track around the football field with a point of interest in the middle. At one point, the point of interest was barely visible in the frame. In my mission settings I have it configured to focus on point of interest and I make sure that each one of my waypoints is set up that way as well. Have any of you had the same problem? Am I doing something wrong? Any suggestions, help or ideas would be appreciated.
Litchi Point of Interest Not Centering Properly
- Thread starter VernCook
- Start date
Post a link here to the mission plan from the hub that exhibits the issue and we can take a look see as to what the issue may be, planning wise.
Thank you. Here is the mission my students did around the school. You'll see a point of interest in the middle of the football field. Towards the end of the orbit of the field, the point of interest was barely in the frame. Mission Hub - Litchi
Not sure that this will make much difference, but you have all of your altitudes set to AGL, but the POI #2 is not. That could be part of the issue. Just a degree or two could make a difference.
I rather doubt that would make much difference when it comes to keeping the point of interest in the center of the frame. It might make a difference when it comes to the gimbal pitch, but the point of interest should still be centered. At least that's my understanding. The kids were the ones who did this mission so there are some mistakes.
I doubt it as well, according to the elevation data there is only a foot or so drop between the end of the track and the POI itself. The same applies to the AGL altitudes. It is quite level actually. You say this is only near one end of the orbit of the track?
Just to throw this out. You didn't mention it, but you didn't happen to lose signal during the mission did you? The POI's will not work properly when you lose signal. It will retain the last gimbal angle that it had before signal loss. I suggested that because it looks like then buildings between waypoint 1 and the field might block the signal from the RC. Just a thought to throw out.
It was toward the end of the orbit when it became the worst and the point of interest was nearly out of the frame. That being said, I have done this when I was first testing the Litchi app and I was orbiting my house. The result was the same - the house wan't always centered in the frame, and sometimes it was quite out of frame. When I did the videos for the golf course, I often found that the green was not centered in the frame by the time I reached the end of the fairway and I would have to move the point of interest off to the side of the green in order to get the shot focused on what I wanted. This kind of behavior forces me to do several test runs before getting a usable video, which of course increases the amount of time I have to spend on a job. I don't think it should be working this way.
I did not lose the signal since the cruise speed was slow enough for us to follow the drone around on its course. And as I have said above, I see this also on other shots. The golf course is a good example. I can fly the drone straight down a fairway with the green as the point of interest and by the time I get to the end of the fairway, the green is no longer centered in the frame. Needless to say, the signal is never lost on those short flights. That signal loss behavior is something I was not aware of, however, and I appreciate your comment. I will definitely keep this in mind in the future.
One other thing. There seems to be some confusion between heading and gimbal pitch. If I understand it correctly, the heading is what keeps the camera pointed at the point of interest horizontally. The gimbal pitch keeps the camera pointed at the point of interest vertically. My problem is with the heading, not the gimbal pitch.
I am really not certain how the headings are calculated with the POI equation in the app or the hub. They are not exact, I will agree with that point. However, you mentioned changing the position of the POI to adjust the heading. What you may need to do is adjust the actual Waypoint heading in the hub or in the app itself by a few degrees, depending on which waypoints are off center from the calculations. If you look visually at for example Waypoint 24, 25 and 26 in this mission the nose of the aircraft is slightly off center at Waypoint 25. That's about the only thing I can think of to correct that.
You are correct. And loss of signal only affects gimbal pitch (and speed changes).
I've had this problem too on a particular mission or two. Over multiple runs. Never fully sorted it out. Problem eventually went away. May have been a compass calibration in between although in theory shouldn't matter because a bad calibration might make the whole bird fly off course. Just guessing on that part.
I think I've come to that conclusion as well. Since I am still in learning mode, I often make stupid mistakes. The other day, for example, I set the drone on a metal surface to launch because it happened to be the flattest place to set it. Big mistake. I got compass errors to the point that the drone wouldn't even take off - lucky for me. After that, I had to recalibrate the compass. I've seen compass errors before but had ignored them. Drone seems to be flying better now.
Are you saying you recalibrated and reran the mission and the heading is now spot on the POI?
I did not rerun that particular mission since I have completed that job and didn't want to have to go out onto the golf course again, but I have flown a couple of test missions since recalibrating and the camera centers on the point of interest much better. It's still not spot on, but it's significantly better and is totally acceptable.
If this statement is correct and acknowledged by Litchi, then it’s a major failure to the SW, because from the very beginning Litchi was meant to perform preloaded missions, in the exact manner the pilot has designed them, no matter if the radio link AC-RC would be interrupted for any reason during the mission, mainly due to increased distance btw AC & RC.
Is Litchi confirming the gimbal angle is not following mission’s parameter if the AC-RC radio link is broken during the mission ? If yes, then what purpose has this strange behaviour? because to me it doesn’t look like a tech limitation, but rather a SW developer choice...
I tend to agree with you. My understanding was that the missions were loaded into the drone so they could be flown by the drone independently from the controller. That way, the mission continued to fly even though contact was lost with the controller.
This is true, however as noted many times on numerous posts, if signal is lost, any pre-planned mission gimbal and speed changes are ignored. Speed and gimbal angles will remain at the last setting prior to signal loss. The remaining parameters will continue to function.
