Question: Who tested the waypoints mission on the Evo 2.

[parkgt]

First and foremost i can edit my flight at home on a big *** screen. A lot of users, mostly young users, are perfectly fine editing stuff on a phone. I am not. Never will be. Second: You can even set the radius of the turns, alt over ground, etc... lots of stuff, just check it out on their page, i don't want to advertise here ;-) but i love litchi.
[/QUOTE

I agree that Litchi and better yet Dronelink are superior than the others. I was wondering why you thought the Autel WP was better than GO4 WP 2.0?

 

[claudius62]

The Perfect Couple, Litchi and Virtual Litchi Mission which sends you in Google Earth pro to do the flight and to be able to refine all angles of shots and focus which results in the perfect mission flight
Of course, you spend a lot of time there, but during the editing, you gain some.

 

[JimDow]

To ensure your smooth, dynamic video I too think that flying a preprogrammed waypoint flight mission is the way to go - especially if you need to reproduce the flight. I've attached a sample (waypoint) flight plan that's effective and efficient. For the most part, you never have to stop, slow down, or drastically change direction. It's not made with circles, splines or Bezier functions; and there are no parallel segments so it demonstrates a much more diverse view perspective throughout. By pulling out one or more trajectories, it can be made to work for a tower or singular point of interest. Isolating a single trajectory segment gives you a constant tangent angle (to the flight path) to focus on a point of interest without ever moving the camera throughout the entire flight path. Maybe one of the drone vendors will get interested in providing some real capability one day.

 

[piperz]

yes, with a can of beer in me i think i can, smooth out the jitters.. but why obsess on the ezact same track, you loose creativity by just programming points. i think your tower needs to be more centered , its a bit laggy, i can fly better on the sticks

I just wanted to step in to say, as a user of both EVO and Mavic platforms, that there is a lot of reasons someone would want to use automation in their recording. For most users, I'm sure manual flying is totally adequate. For those users who are in a professional field, they often require more complexity in their recordings. Some examples to consider would be a situation where you would like to automate the drones flight path so you can dedicate all of your attention to the movements of the camera. In TV / Film production sometimes you need to do the same shot multiple times, so being able to have the task automated insures that all takes will have close to the exact positioning. In TV Production you often do several takes of a sequence using stand-in's before you film your sequence with your primary actors. Again, this insures that the path you chose for the setup shot, will be the exact same you use for the final sequence. Sometimes you have a complex flight routine that has areas where it disconnects from the remote, having a mission path setup gives you the peace of mind that your film sequence isn't going to be ruined by a signal drop. This example in particular is the scenario that I used mission automation the most for.

 

[Shinespal]

First and foremost i can edit my flight at home on a big *** screen. A lot of users, mostly young users, are perfectly fine editing stuff on a phone. I am not. Never will be. Second: You can even set the radius of the turns, alt over ground, etc... lots of stuff, just check it out on their page, i don't want to advertise here ;-) but i love litchi.

EVO 2" data-source="post: 53781" class="bbCodeBlock bbCodeBlock--expandable bbCodeBlock--quote js-expandWatch">

First let me say I made a mistake the video I posted was untouched from the EVO 2 but not the mission flight. Sorry....Ok ...The behavior was fine.....I had 9 points I programed the all a little different...Speed ....Altitude and gimbal pitch..there was no gradual change it was fast and sharp ...yes by it self

 

[piperz]

Yes, but what is his behavior on mission.
Gradual change in altitude between waypoints?
Is the sequence between waypoints curved or does it turn quickly on itself?
It is not very visible on the video because there are a lot of stops.

Here is an automated waypoint mission (No user control)

 

[claudius62]

One cannot judge on such a course which includes only three or four waypoints and no change in altitude.
All I saw was a slowdown at the waypoint where there is a change of course to go back.
Gradual slowing down of speed like the Evo 1 (this is normal since it is a Px4 platform common to both) from the initial speed to around 2 km / h then gradual acceleration up to the programmed speed. This (bad) transition unfortunately increases the flight time by 10 seconds at each waypoint. And the same time for a change in altitude.
On a light route of a few waypoints, it is not transcendent, it will suffice for the assembly to make a connection before and after slowing down, but it is restrictive all the same.
But on a long distance course, with for example 30 waypoints, this will lead to an addition of estimated time for this journey.
28 transition waypoints (I am not counting the departure and arrival waypoint) will lead to a time increase of 280 seconds or 4 minutes 40 to be taken in addition to the time of your mission which is given. It is not negligible if the estimated time is already right for the mission.

 

[FredC]

Having a resource like Litchi sounds like it would be great for cinematic shots. Unless you're really great at flying while moving the gimbal, hitting your shoot points could be tough. Imagine circling a bridge and then dropping altitude, raising the pitch on the gimbal just as you get to a point just above the roadway to look straight down the road. How many times will it take to get that manually? If I could program that shot in, and get smooth transitions between waypoints, I'd get that.

 

[crusho]

I can see wanting to always be at the controls, especially for what might get replaced with the waypoint mission, but as I've read through this thread, it's clearly not possible. As someone who can envision a shot in my head (let's say not of a moving object, but through a city or along a complex shoreline), it would be great to offload most of the piloting to my "virtual pilot" while I focus on being the cameraman. But where I envision a smooth, sweeping 90-degree turn while descending 75 feet, instead I get a stop, a pause, a 90-degree jerk in course that's so hard as to be painful, and finally a rapid decent before even moving on to the next waypoint. That's great for a grid pattern, shooting a series of photogrammetric images for Drone Deploy et al — yet worthless for the cinematic video footage I described above. The only point for an Eo II waypoint mission as it's currently implemented is to get the drone to a specific location — a single A to B — without having to hold any sticks down. This sucks, as it's pointless, an I hoped the Evo II to be my cinematic/photography drone; I'd use my Skydio 2 for mapping or inspections.

So my actual question: Everyone talks about Litchi; is it capable of a smooth waypoint mission (with a compatible drone) or is it just another Drone Deploy/etc? I've got a "cinematic/photography" drone (at least when flown manually, and flying even just the basic shot I described isn't a no brainer — it's piloting that takes years of practice, which I'll guess I'll eventually get, but not in the urban area I envisioned that shot, and not before many, many hours — a good thing for sure) in the Evo II, a sport-shooting (and surprisingly commercial work-related within months) with the Skydio, and one just to have the experience of flight (and probably the moments before hitting the ground at full-speed — eventually — I haven't even gotten to the sim yet. Don't drink and drone shop lol) in the DJI FPV.

If someone says yes to my Litchi question (which I'm pretty sure is yes), I'm afraid of ending up the Jay Leno of drones. But to answer many rhetorical and actual questions above about "who's flying" and "why wouldn't you want to be on the sticks", it's because some us have fully-realized cinematographic ideas, but it they would be **** if self-piloted. I think the notion of a smart virtual pilot has a lot of uses, and isn't a cheat especially for less experienced pilots. There are some things a computer/AI can do better than any human, and some cinematographic shots are one of them. Can you correct/alter on the fly? Shoot moving objects? No and no, and there's 100 other scenarios where the answer is no, so great pilots will always be needed. At the same time, I could devise a flight path in advance that probably very few, if any human pilot could execute exactly as envisioned/planned out.

If I did get something like an Air 2s for this, it limits my need of the Evo II to a backup and geofencing evasion. I can't even talk about the camera anymore, esp if the 2s actually has a true 1-inch sensor. Which sucks, given the amount sunk into it. Is the Evo missing something in hardware or firmware that keeps it from being/becoming Litchi-compatible?

 

[Rattlebars]

The latest updates to the firmware make the Evo gradually change altitude. It does pause at each waypoint. One way to focus a shot is to drop a POI or two. Here's a video showing a construction site I'm keeping progress reports on. The mission has 2 POIs and shows the gradual flight altitude behavior. Actually, I cannot believe that anyone can do cinematic shots with no post production on any drone. Research litchi there are ways to "shape" alititude changes but as of yet, we cannot import KML effectively.

 

[kenautelevo2pro]

^I like the gradual altitude change (as opposed to rising straight up at the waypoint before proceeding).

I guess you can exert a fair amount of control over the altitude change by using multiple waypoints. as my post production skills improve, I am finding the few limitations on missions to be more tolerable.

 

[crusho]

The latest updates to the firmware make the Evo gradually change altitude. It does pause at each waypoint. One way to focus a shot is to drop a POI or two. Here's a video showing a construction site I'm keeping progress reports on. The mission has 2 POIs and shows the gradual flight altitude behavior. Actually, I cannot believe that anyone can do cinematic shots with no post production on any drone. Research litchi there are ways to "shape" alititude changes but as of yet, we cannot import KML effectively.

Well of course there's post, but if you could pre-program a route — even just a portion, esp one where you really needed to focus on the camera — it's gonna make work in post that much easier.