iOS Waypoint App Comparison: Litchi 1.0 vs Autopilot 3.0

[beazergood]

@2nd2non I am going through the process of getting my P3P ready for its maiden flight.
I'm really excited by the capability Autopilot has and really appreciate the info in this thread!
Quick question, do you recommend getting used to how waypoints work in DJI go before venturing into Autopilot?

 

[KeithS]

Im a huge user of Litchi and i cant fault any aspect of it at all.
However i am on Android.
Ive planned some very complex and intricate missions and never been in a sweat once about it coming back.
I often push it beyond signal range, but am surprised that when i do use it the signal range is further than using the Dji Go app in the same area.
Being able to plan your flight via any web browser and instantly have it linked to your mobile device is so convenient.

 

[2nd2non]

@2nd2non I am going through the process of getting my P3P ready for its maiden flight.
I'm really excited by the capability Autopilot has and really appreciate the info in this thread!
Quick question, do you recommend getting used to how waypoints work in DJI go before venturing into Autopilot?

Waypoints for DJI GO is quite different in its implementation than autopilot. If this is your first time flying a drone with waypoints, DJI GO is a good safe place to start since you have to manually fly each waypoint first. This will help give you perspective (height and distance perspective). Autopilot also has ability to manually fly and register waypoints. So you could progress here after reading the flight school documentation for the mode. Next step could be to pre-configure waypoints from the ground then run the mission. Take it slow and test in an open field. Next thing you know, you'll be able to venture out with much confidence!

 

[2nd2non]

Im a huge user of Litchi and i cant fault any aspect of it at all. I often push it beyond signal range, but am surprised that when i do use it the signal range is further than using the Dji Go app in the same area.

Litchi allows you to set the video channel up to 32 channels. DJI GO can do this if you use the 32 channel hack. The higher channels usually do not have much interference.

Being able to plan your flight via any web browser and instantly have it linked to your mobile device is so convenient.

yes it is. Have you tried setting it up with short distances (ie. 3 meters) then capturing Timelapse or long exposure photos? Check out this using Litchi ->

Here's where autopilot puts this concept on steroids-> same concept of long exposure Timelapse photos, but putting it into hyper lapse with their continuous photo mode ->

 

[Yeager74]

The issue is actually present in all Modes in Autopilot not just Waypoint, and several Modes in other software packages as others have pointed out.

At any rate, your conclusion seems a bit extreme. Obviously you are entitled to you own opinion, but as we stated previously, we have never actually experienced the RTH issue in real world testing (thousands of flights) unless we intentionally made it happen (by turning off the RC). Even in the face of this data, it is completely understandable why you don't want to risk it, but it seems like just using the other options in the meantime is your best bet*. Not entirely sure why Autopilot has to be forever written-off.

DJI may be slow to produce an update, but they will produce an update if they say so. This has always been our experience with them in the past and we have no reason to believe this time is different. Even still, we do have another inquiry into them as to when the fix is schedule (i.e. which firmware version).

*On using the other options - I assume the reason you are so adamant about this issue and active in this thread is because you would actually prefer to use Autopilot. What is it about Autopilot that appeals to you over the other options? In other words, is it one or more of the unique features that we offer because we don't support LCMC, or is it something else?

Thanks for your reply.

The pre-programming of constraints (individual WP speed and altitude) in the waypoint mode of Autopilot are great and unique as a combination. The logical (behaviour) build into how the plane meets these constraints are great. As seen from an ("semi")autonomous perspective (besides from not having the LCMC options, for whatever reasons) the other known iOS applications does not provide the same level of operator detailed input and control. It's understood that the current Autopilot interface is unique and not able to incorporate LCMC, which I personally do not care to much about (though having it as an option would be a great, but not compromising the "unique" features of AP).

There's no reason why Autopilot could not be picked up from the bin at a later stage, when/if DJI actually does provide a FW fixing the current bug, which you correctly acknowledge is all about risk management, which is why I stand by my conclusion until then. As mentioned, I do perform beyond LOS (with DBS itelite) and I need the be confident that the "drone" will failsafe RTH.


P3P W323B GL300C, APP V.2.7.0, AIRCRAFT V.1.6.40, RC V.1.5.70, AUTOPILOT V3.2A.

Sent from my iPhone using PhantomPilots mobile app

 

[Hangar]

when/if DJI actually does provide a FW fixing the current bug

Just received word from DJI explaining why the recent firmware release doesn't have the fix:

Yesterday’s firmware release was only related to a camera change to meet a regional regulation. We are still trying to get the ETA for the full firmware release with the flight controller (and everything else) upgraded.​

 

[With The Birds]

Just received word from DJI explaining why the recent firmware release doesn't have the fix:

Yesterday’s firmware release was only related to a camera change to meet a regional regulation. We are still trying to get the ETA for the full firmware release with the flight controller (and everything else) upgraded.​

I reckon i will be able to perform a perfect orbit with fixed poi in DJI Go before dji gets this firmware out. I will be waiting though, i have a few AP missions ready to run.

Btw- given autopilot with the current SDK will attempt RTH at critical battery in aignal loss condition if I set the critical to 30% will AP initiate RTH at that point?

Could be a good interim work around.

 

[Hangar]

if I set the critical to 30% will AP initiate RTH at that point?

Out testing shows yes.

 

[Yeager74]

Out testing shows yes.

Really? Well, that would be an acceptable way of working around the current bug, in my opinion. My understanding was that, when reaching the DJI GO user selectable value of the "critical battery warning", the plane would execute auto-land and NOT execute RTH. I thought I had tested this already, with a different result (not RTH, but auto-land). I could be wrong on this!

Let me clarify the test at hand;

1. DJI GO "Critical Battery Warning" user setting set at 30%. (Ps. NOT talking about "low battery warning" (10 secs and no operator input etc.)).
2. Autopilot application, mission mode.
3. RC Signal loss (simulated by turning off RC)
4. @ 30% Battery Level = critical battery RTH execution as opposed to auto-land.

Is that the current assumption/question?

I will test this on Saturday (Hong Kong time), if wx permits.




P3P W323B GL300C, APP V.2.7.0, AIRCRAFT V.1.6.40, RC V.1.5.70, AUTOPILOT V3.2A.

Sent from my iPhone using PhantomPilots mobile app

 

[Yeager74]

Just received word from DJI explaining why the recent firmware release doesn't have the fix:

Yesterday’s firmware release was only related to a camera change to meet a regional regulation. We are still trying to get the ETA for the full firmware release with the flight controller (and everything else) upgraded.​

Thank you for this updated information.


P3P W323B GL300C, APP V.2.7.0, AIRCRAFT V.1.6.40, RC V.1.5.70, AUTOPILOT V3.2A.

Sent from my iPhone using PhantomPilots mobile app

 

[Hangar]

Is that the current assumption/question?

Yes.

 

[Roamer105]

No one is forcing anyone to do anything. If you want LCMC, use Litchi or whatever tool is best suited for that mission. When you want to do things that only Autopilot can do, use Autopilot. It's that simple.

Exactly!
I recently flew a mission that forced me to use (and learn about) Litchi. The mission was to fly around the back and over a small (700 foot high) volcano with a max altitude gain of 850 feet over home point and max distance of 1 mile to the RC while behind the mountain. Without trying to do this with Autopilot I knew this would not work as I was sure to lose connection at some point when the quad would be behind the mountain. Litchi flew the mission perfectly, however Autopilot would have done better smoothing the flight path. I am always worried to lose connection when flying larger missions with Autopilot. I prefer to set the mode to "hover" on loss giving me time to regain control rather than return-to-home while not in control. Unless the RTH altitude is very high, there are few guarantees that RTH would not end up crashing into a hillside.

Clearly, I am a BIG supporter of Autopilot. Airspace is fantastic. I have been flying specialized follow missions (following a working search dog with iPhone 6 attached to his harness at relatively low altitude (30 feet and short distance 50 feet)) totally unattended with confidence many times, always without problems and excellent results and NO FLY-OFFS! Most importantly, the P3P maintains RELATIVE altitude to the iPhone 6! So, even if the dog runs up a hill, the follow footage is great and the mission is safe.

One feature I REALLY liked a lot with Litchi is the ability to fully develop a mission off-line (on a real computer) using mission hub and immediate cloud sharing to all devices. Mission hub provides the relative altitude difference of any waypoint to the first waypoint. This helps tremendously when developing complex missions in mountainous terrain. So THIS would be a great feature to add to Autopilot in addition to a possible choice for an LCMC (even if degraded capabilities) option.