Litchi waypoint run stops in mid-mission

[deggers]

Interesting. Assuming I am reading @BudWalker 's new fields correctly, on the first lap it looks like the wind peaked at ~ 15mph. Second lap when you ground to a halt it pretty much doubled in strength, increasing all the way up to 33mph (which corresponds with max speed in waypoint mode) and on the final lap it abated back to around 15mph again.

Cheers for looking into it - out of interest and for future ref - how do we find the recorded windspeed in the files please?

 

[BudWalker]

Cheers for looking into it - out of interest and for future ref - how do we find the recorded windspeed in the files please?

The wind speed calcs aren't well enough understood to provide an easy answer. I mentioned in this post
Litchi waypoint run stops in mid-mission
there is a beta version of CsvView that contains those wind calcs. All that's really known are the labels that DJI gives to those values. I think the AirSpeedBody values are pretty clear. But, the WindSpeed values seem to be inconsistent. I've done some experiments where I send up my P3 and put it in ATTI mode to measure the winds aloft. Then follow that with Mavic and see if I can make sense of the wind values. So far, I'm still kinda confused. I was hoping that making this available to others before I understood it would help create some understanding.

If the AC is hovering then air speed should be the same as wind speed. But, they aren't here at time 260.926 where velH = 0.1, wind speed is 13.8, but air speed = 10.49 (airSpeedBody:X = 8, airSpeedBody:Y =6.8)

Anyway, in your flight there weren't any obstacles seen by the front sensor. But, it did indicate OA:emergencyBrake at 244.788. Is it possible that the emergency brake gets turned on in waypoint missions when the air speed has to exceed a certain value?

 

[deggers]

@BudWalker I've been chatting to someone Litchi's facebook page and asked them the question. I'm guessing round about this point your referencing where it says after pausing (to assess whats going on) I resume but then goes into an emergency break. At least we know it wasn't object avoidance kicking in then.

244.283 : 1718027935 : 18753 [L-RC]rc cmd:STD_PAUSE_CUR_MIS
244.285 : 1718036495 : 18753 [L-FLYMODE]CTRL rc resume waypoint_mission
244.905 : 1720826740 : 18784 [L-RC]Emergency break. rc vib!

However the person that was replying from Litchi was "guessing" what log results meant last night

• received idle vel cmd 8.999998 from app
• ack idle vel 0 to app

"if I were to guess the 0 here is probably a result code, where 0 means success" - Litchi

So ack idle vel 0 to app is an accept speed instruction reply. Would make sense from the logs.
8.999998 m/s being 20.1mph (the set cruise speed)

I'll let you know what they say re asking what would cause an emergency brake without object avoidance being invoked.

 

[deggers]

@Logger If it received a 33mph gust and the logs verify it was getting instructions to fly @ 20.1mph - surely it would have shifted backwards or off line substantially instead of (kinda) hovering? My Litchi max speed was set at 33.6mph (max) but this is only if I over-ride on the forward stick with object avoidance OFF - as object avoidance on restricts it to 20 something mph? - so its not as though it ramped itself up to 30mph to keep itself in some kind of situ? I didn't push the forward stick just rotated to craft towards me away from the reservoir (incase of sun/reflections stopping it moving)

I guess what I'm getting at is if a gust of wind is overpowering it by 13mph difference - wouldn't it move "somewhere" at 19ft/second - which would have shown on the track and it was hovering near enough in place for some considerable time if 30mph gusts where hitting it.

Sorry for so many questions btw.

 

[Logger]

@Logger If it received a 33mph gust and the logs verify it was getting instructions to fly @ 20.1mph - surely it would have shifted backwards or off line substantially instead of (kinda) hovering? My Litchi max speed was set at 33.6mph (max) but this is only if I over-ride on the forward stick with object avoidance OFF - as object avoidance on restricts it to 20 something mph? - so its not as though it ramped itself up to 30mph to keep itself in some kind of situ? I didn't push the forward stick just rotated to craft towards me away from the reservoir (incase of sun/reflections stopping it moving)

I guess what I'm getting at is if a gust of wind is overpowering it by 13mph difference - wouldn't it move "somewhere" at 19ft/second - which would have shown on the track and it was hovering near enough in place for some considerable time if 30mph gusts where hitting it.

Sorry for so many questions btw.

The speed you command in waypoint capable programs such as Litchi is GPS ground speed and not airspeed. So if you program 20mph it attempts to maintain this groundspeed as you fly upwind (up to a point) by increasing airspeed above programmed speed or downwind by decreasing it. I imagine it will increase pitch into a headwind until it reaches the maximum WPT mode pitch for still air which equates to 15m/s or 33mph (regardless of your programmed max GPS speed). It is doing this in an attempt to maintain your programmed 20mph.
The Litchi max 33.6 is the max GS it will allow itself to fly in WPT program mode which appears common to all apps.

 

[Logger]

The wind speed calcs aren't well enough understood to provide an easy answer. <snip>

If the AC is hovering then air speed should be the same as wind speed. But, they aren't here at time 260.926 where velH = 0.1, wind speed is 13.8, but air speed = 10.49 (airSpeedBody:X = 8, airSpeedBody:Y =6.8)

Is the aircraft hovering. Perhaps my CSV is timed differently, but at 290.969 I see vel H 1.9 ??

 

[BudWalker]

Is the aircraft hovering. Perhaps my CSV is timed differently, but at 290.969 I see vel H 1.9 ??

I also see vel H 1.9 at 290.969. But, I believe I said "at time 260.926 where velH = 0.1"

 

[Logger]

So you did. I am Lysdexic again Sorry..
Might try comparing my DAT derived wind values with some Aircalc derives and see how they compare.

 

[BudWalker]

So you did. I am Lysdexic again Sorry..
Might try comparing my DAT derived wind values with some Aircalc derives and see how they compare.

Yes, I've also had problems with Dysexlia.

Maybe we should start another thread to talk about the wind calc issues.

 

[BudWalker]

@BudWalker I've been chatting to someone Litchi's facebook page and asked them the question. I'm guessing round about this point your referencing where it says after pausing (to assess whats going on) I resume but then goes into an emergency break. At least we know it wasn't object avoidance kicking in then.

244.283 : 1718027935 : 18753 [L-RC]rc cmd:STD_PAUSE_CUR_MIS
244.285 : 1718036495 : 18753 [L-FLYMODE]CTRL rc resume waypoint_mission
244.905 : 1720826740 : 18784 [L-RC]Emergency break. rc vib!

However the person that was replying from Litchi was "guessing" what log results meant last night

• received idle vel cmd 8.999998 from app
• ack idle vel 0 to app

"if I were to guess the 0 here is probably a result code, where 0 means success" - Litchi

So ack idle vel 0 to app is an accept speed instruction reply. Would make sense from the logs.
8.999998 m/s being 20.1mph (the set cruise speed)

I'll let you know what they say re asking what would cause an emergency brake without object avoidance being invoked.

Hard to tell what the eventLog stream is actually saying. The nomenclature changes with every firmware update. Another interpretation is that the Mavic is pausing the mission because of wind speed. Before the pause the roll and pitch angles are pretty high and only producing a vel:H of 1.7. After the pause the roll and pitch angles are only slightly reduced producing a vel:H of 0.0. That suggests that the wind speed is pretty high.

 

[deggers]

Must say this info you guys can put together is brilliant - really impressed.

So we can conclude, thanks to you guys, that the mav stopped because it was getting a battering by the wind and as I thought it had stopped/was struggling didn't think it was the wind but worrying about the sun obj detection I paused the mission then resumed and it eventually came back home.

I asked Litchi re the emergency brake on and they said that the dji firmware controls this so not something they can answer themselves. Perhaps one for the DJI forum if I wish to investigate further.

On I side note I went out with the mav today and did over 4 miles with it on Litchi missions and no issues at all.

Just found I can make airdata shared for the flight so I will leave it below but you can see the amps hitting purple also on that leg.

Airdata UAV - Flight Data Analysis for Drones

Thanks again.