Labroides Posted at 4-10 22:54
And you still haven't explained what you think I got wrong.

What I was giving up on, was hoping too get a clear answer or explanation from you.

nah, guess you do not want to understand

if distance to homepoint does not change durring autoland, up to the moment that the height is zero than  the simple conclusion is that this drone did not move forward at all.

it is not chinese....isn`t it ?

But i leave it here, explanation is easy to understand in my previous posts.

Irrelevant data, nah it is more data to make the picture complete. (cam move for examople from down to zero during descend).

cheers
JJB

JJB* Posted at 4-11 00:30
nah, guess you do not want to understand

if distance to homepoint does not change durring autoland, up to the moment that the height is zero than  the simple conclusion is that this drone did not move forward at all.

nah, guess you do not want to understand
I've been begging you to explain, but you've ignored the requests.

if distance to homepoint does not change durring autoland, up to the moment that the height is zero than  the simple conclusion is that this drone did not move forward at all.
If so, that would be correct. ... up to the point that the drone was very close to the ground, but the video clearly shows the drone moving quickly along the ground after it touches the ground.
It's not a "bounce"

Now explain what these numbers in red mean and how they fit with the fiction you've been insisting on..
And an apology would be nice too.

it is not chinese....isn`t it ?
More like double dutch.

Labroides Posted at 4-11 00:57
nah, guess you do not want to understand
I've been begging you to explain, but you've ignored the requests.

Now explain what these numbers in red mean and how they fit with the fiction you've been insisting on..


JJB* did explain those that you've highlighted in red, in reply #41, what with your super intelligence, you should have picked up on it, anyway, he said...


''so all this make me think, with the text of the operator, that drone indeed descende to fast, 'hit; the ground and moved bit forward.''

Not sure why you would want an apology from him though, to me it looks like it should be you doing the apologising.

Bashy Posted at 4-11 03:28
Now explain what these numbers in red mean and how they fit with the fiction you've been insisting on..

Thankyou for your contribution to science.It's always so good to have your insight into things technical.

But if you had an IQ above room temperature you'd realise that a drone descending too fast (which the data shows DID NOT HAPPEN, any bounce would be upwards, not at 90° to the direction of motion.
And I know it's only a trivial detail, but the numbers in red are the distance from the homepoint.
He didn't explain them - in his last post he's inexplicably suggesting that the drone did not move forward at all.

Now go away and chase a squirrel or something.

It's just a guess, but after watching video and logs, it looks to me like the home point is close to the wall edge with a bush on it.
The question is how tall is the bush with the thin branches on it. If the height for slow down for the landing is lower than the height of the plant, what is shown in the video will happen.
So I think the drone hits the thin branches that sent the drone forward before the landing speed is reduced.

Aurgelme Posted at 4-11 04:58
It's just a guess, but after watching video and logs, it looks to me like the home point is close to the wall edge with a bush on it.
The question is how tall is the bush with the thin branches on it. If the height for slow down for the landing is lower than the height of the plant, what is shown in the video will happen.
So I think the drone hits the thin branches that sent the drone forward before the landing speed is reduced.

Despite what some people have said, the landing speed did slow down.
The data shows this very clearly.

Labroides Posted at 4-11 05:01
Despite what some people have said, the landing speed did slow down.
The data shows this very clearly.

Yes, I can definitely see the speed going down, but not the last bit of the landing where it hits something that sends it forward.

Aurgelme Posted at 4-11 05:11
Yes, I can definitely see the speed going down, but not the last bit of the landing where it hits something that sends it forward.

Just look at the tables of numbers I posted (twice).
They show that the drone slowed exactly as it is supposed to on descent before touching the ground.

I watched the video. The aircraft hit the bushes or the aerodynamics changed due to the bushes.
I once landed on a very narrow ledge, and when the drone was about to touch the surface, it was pulled hard to the side and stopped at the edge. Those. a strong influence on the aerodynamics of what is below.
But it is not clear from the video why the rate of descent was such. But it was also the pilot's mistake that he did not take off from the center or did not land in manual mode in the center of the site.

Drone.Hunter Posted at 4-11 05:25
I watched the video. The aircraft hit the bushes or the aerodynamics changed due to the bushes.
I once landed on a very narrow ledge, and when the drone was about to touch the surface, it was pulled hard to the side and stopped at the edge. Those. a strong influence on the aerodynamics of what is below.
But it is not clear from the video why the rate of descent was such. But it was also the pilot's mistake that he did not take off from the center or did not land in manual mode in the center of the site.

But it is not clear from the video why the rate of descent was such.
But the data clearly shows that there was nothing wrong with the descent rate.
It did slow just as it is supposed to.

Drone.Hunter Posted at 4-11 05:25
I watched the video. The aircraft hit the bushes or the aerodynamics changed due to the bushes.
I once landed on a very narrow ledge, and when the drone was about to touch the surface, it was pulled hard to the side and stopped at the edge. Those. a strong influence on the aerodynamics of what is below.
But it is not clear from the video why the rate of descent was such. But it was also the pilot's mistake that he did not take off from the center or did not land in manual mode in the center of the site.

For what it's worth, I did take off from the centre of the garden, roughly 2 metres in front of where the RTH/auto land decided to land the drone. So this was another factor leading to what happened.

I usually do land in manual mode and hand catch with the Mavic Mini I owned previously. But as mentioned by the posters in comments #4, #6, and #22, DJI drones usually slow down (more than it did in this case) about 2m above the ground, giving more of a chance to cancel the auto landing.

Anyway yes going forward I will definitely not be relying on the accuracy of RTH for auto landing.

KingCon Posted at 4-11 05:59
For what it's worth, I did take off from the centre of the garden, roughly 2 metres in front of where the RTH/auto land decided to land the drone. So this was another factor leading to what happened.

I usually do land in manual mode and hand catch with the Mavic Mini I owned previously. But as mentioned by the posters in comments #4, #6, and #22, DJI drones usually slow down (more than it did in this case) about 2m above the ground, giving more of a chance to cancel the auto landing.

I have after takeoff, the point of the house has shifted a few meters. In dangerous places, landing is only in manual mode, but you can also fly to the house in automatic mode.

KingCon Posted at 4-11 05:59
For what it's worth, I did take off from the centre of the garden, roughly 2 metres in front of where the RTH/auto land decided to land the drone. So this was another factor leading to what happened.

I usually do land in manual mode and hand catch with the Mavic Mini I owned previously. But as mentioned by the posters in comments #4, #6, and #22, DJI drones usually slow down (more than it did in this case) about 2m above the ground, giving more of a chance to cancel the auto landing.

For what it's worth, I did take off from the centre of the garden, roughly 2 metres in front of where the RTH/auto land decided to land the drone.
Two metres isn't enough clearance to be certain with RTH autolanding because consumer GPS accuracy is around +/- 3 metres.
Most of the time it will be better, but sometimes it could be even worse.
For this reason you need to ensure a good clear area around the launch point if the drone is likely to be coming home on its own.

But as mentioned by the posters in comments #4, #6, and #22, DJI drones usually slow down (more than it did in this case) about 2m above the ground, giving more of a chance to cancel the auto landing.
I've pointed out many times in this thread, your drone did slow down as it should.
Read my first post where I explained this in detail and showed the data to support this.

KingCon Posted at 4-11 05:59
For what it's worth, I did take off from the centre of the garden, roughly 2 metres in front of where the RTH/auto land decided to land the drone. So this was another factor leading to what happened.

I usually do land in manual mode and hand catch with the Mavic Mini I owned previously. But as mentioned by the posters in comments #4, #6, and #22, DJI drones usually slow down (more than it did in this case) about 2m above the ground, giving more of a chance to cancel the auto landing.

Just for reference, unless the takeoff / landing site is out in the open I very rarely let a  drone land automatically and that includes a Mavic 2 Pro with precision landing (which has proven cm perfect ).
Even if I do let it land automatically I watch the drone and am ready to give joystick commands if I think something is awry.

I am not knocking you, we all have to learn, but I think not using throttle to slow the descent to what is 'normal' for the drone when the drone was low was your big mistake, it would have given you time to cancel the RTH landing and take full manual control.

I have actually made a Mini 1 or 2 climb during the landing phase, just as a test.

If possible MAKE SURE OA is off during landing.

Labroides Posted at 4-11 04:42
Thankyou for your contribution to science.It's always so good to have your insight into things technical.

But if you had an IQ above room temperature you'd realise that a drone descending too fast (which the data shows DID NOT HAPPEN, any bounce would be upwards, not at 90° to the direction of motion.

I love it when you abuse folk, it really brings out your intelligence in your posts lol

You asked JJB* and i answered for him, the question and answer is right there in my reply lol

Bashy Posted at 4-11 06:36
I love it when you abuse folk, it really brings out your intelligence in your posts lol

You asked JJB* and i answered for him, the question and answer is right there in my reply lol

You really are the dumbest person in this forum.
You couldn't understand a single thing I've written here.
There are two many multi-syllable words for you to deal with.

Sean-bumble-bee Posted at 4-11 06:33
Just for reference, unless the takeoff / landing site is out in the open I very rarely let a  drone land automatically and that includes a Mavic 2 Pro with precision landing (which has proven cm perfect ).
Even if I do let it land automatically I watch the drone and am ready to give joystick commands if I think something is awry.

I think not using throttle to slow the descent to what is 'normal' for the drone when the drone was low was your big mistake,
Is every post I made in this thread invisible?
It seems that no-one has read any of them.

Labroides Posted at 4-11 00:57
nah, guess you do not want to understand
I've been begging you to explain, but you've ignored the requests.

Indeed RED where height is zero, so touching ground before moving forward.
Thanks Labroides to make my point solid using your chart.

• I would suggest you check some other MINI3PRO logs for how this drone adjust it speed during landing.
• Do you see more logs where the cam is still moving from down to zero pitch so close to the ground?  (why, bit fast descend and too late to stop really)
• Plus understand what the X and Y speed component means in the log and why they are zero in this log until sensor height became zero and drone bumped forward.
• Plus if you cannot use the correct reply, don`t start with double Dutch ect.  Its a bit childish to stop using arguments ; just writing "i don`t understand you...."
• Plus  be friendly to all other users on this forum, if you cannot beat them... no need for calling names or be disrespectfull to others.
  

cheers
JJB

KingCon Posted at 4-11 05:59
For what it's worth, I did take off from the centre of the garden, roughly 2 metres in front of where the RTH/auto land decided to land the drone. So this was another factor leading to what happened.

I usually do land in manual mode and hand catch with the Mavic Mini I owned previously. But as mentioned by the posters in comments #4, #6, and #22, DJI drones usually slow down (more than it did in this case) about 2m above the ground, giving more of a chance to cancel the auto landing.

Hi KingCon,

Your HP may be 2 meters away from the initial take-off point, buit has nothing to do with you landing.
At the end of RTH into autolanding your drone decended down above the HomePoint.
But at start of your autolanding (so direct above HP) you gave some Roll input to the left.This means you cannot say that your say that your drone was 2 meters offset from HP (as a result of rth+autolanding).


There is a misunderstanding about GPS accurancy.
Suppose you have a garden in the shape if a circle, diameter 15 meter.
If you takeoff from the center of your garden, than HP is set in the middle of your garden ofcourse.
Are DJI drones accurate enough for landing in the middle of the garden on a RTH / Autolanding.
Anwser is YES, always exactly in the middle?  nope, few cm to 50 cm max in my experience on many many drone flights.

BUT if you plot the HP in GoogleEarth then you wil see the GPS accurancy, HP is sometimes few meters off your garden center.
Bad for drone fllying ?  ofcourse not as the HP recorded is the place to return to, not to the HP positioned on a map.

Above only if you take off in a good GPS reception, and during flight landing this is not changed.
In your flight with > satellites and por GPS 'health' always a 5. So no problem for that.


cheers
JJB

Sean-bumble-bee Posted at 4-11 06:33
Just for reference, unless the takeoff / landing site is out in the open I very rarely let a  drone land automatically and that includes a Mavic 2 Pro with precision landing (which has proven cm perfect ).
Even if I do let it land automatically I watch the drone and am ready to give joystick commands if I think something is awry.

HI Sean,

why OA off during landing ??

OA does not steer drone away from an obstacle wich come in view while descending.
Just try your self, autoland few cm in front off an onstacle....  it will land normally or sometimes it will cancel the landing if Landing Protection is doing good work.

Landing protection by use of botton vision sensor.

cheers
JJB

JJB* Posted at 4-11 09:26
HI Sean,

why OA off during landing ??

OA off during landing because it can detect objects and prevent movement, which can cause problems. With the Mavic 2, DJI switch OA off automatically during automated landings, page 19 of the manual "Vision Systems are disabled during landing. Be sure to land the aircraft with caution.".
As far was I remember some of the fly app using drones have a similar line in their manuals.

As far as testing it, no thanks. With a Mavic 2 Pro I accidentally had OA on whilst low and at the home point, I nearly lost the drone as a consequence of it refusing to move.

Switching OA off does not disable landing protection

Labroides Posted at 4-11 06:19
For what it's worth, I did take off from the centre of the garden, roughly 2 metres in front of where the RTH/auto land decided to land the drone.
Two metres isn't enough clearance to be certain with RTH autolanding because consumer GPS accuracy is around +/- 3 metres.
Most of the time it will be better, but sometimes it could be even worse.

Thanks for your inputs, I've had a chance to read over everything properly now.

"Your drone did slow down as it should"
You're really saying my drone slowed down to what you would expect for a gentle auto landing?

Please compare my footage to the below examples of a "normal" auto landing:
https://www.youtube.com/shorts/6PjJ-XxFT5w
https://www.youtube.com/shorts/XJuecWTv99Y

My issue is I don't believe the "Landing Protection" worked anywhere near as well as described on page 18 of the DJI manual - https://dl.djicdn.com/downloads/ ... _manual_v1.0_en.pdf

"Once Landing Protection determines that the ground is suitable for landing, the aircraft will land gently."
So Landing Protection did determine the ground was suitable for landing and proceeded, but the landing was anything but gentle.

"If the ground is determined unsuitable for landing, the aircraft will hover and wait for pilot confirmation."
If there was anything that would have caused this area to be considered unsafe for landing, then it should have given the "Unsafe to Land" warning and prompted me to land manually right? I would have been very happy if it behaved this way, instead of what actually transpired..

What's your thoughts?

KingCon Posted at 4-11 13:29
Thanks for your inputs, I've had a chance to read over everything properly now.

"Your drone did slow down as it should"

What's your thoughts?
I explained my thoughts carefully and clearly in my first post in thins thread.
According to the data, the drone did slow as it should have, but it shot forward as it touched the ground.
What caused that is a mystery.
GPS does not have the precision to rely on for a landing in a tight spot like your garden.

KingCon Posted at 4-11 13:29
Thanks for your inputs, I've had a chance to read over everything properly now.

"Your drone did slow down as it should"

Incidentally I would consider getting a landing pad or hand catching/launching rather than launching or landing on naked soil, gravel to sand but read up on hand launching before you try it with a mini 3.

KingCon Posted at 4-11 13:29
Thanks for your inputs, I've had a chance to read over everything properly now.

"Your drone did slow down as it should"

Hi KingCon,

a) You saw your craft landing to fast....and bc of this touch down moved forward.
b) Data in the log shows that the decend rate in autolanding started at a too low height, bc of this your craft landed to hard.

I also compared your log to many other MINI3Pro logs, yes   to hard landing. And yes, at the end the descend was reduced  but as said  little too late.

In a normal autolanding the Landing protection function, using the downward vision sensor, will check the surface below. If area is not suitable fro landing it will cancel the landing. This can happen at any height during the autolanding. If area is safe to land, drone will reduce speed such that is it almost zero at 0.5 meter above the ground, after that it will land gently.
NOT in your log.

Why?  bc Landing Protection did not identify your landing spot as unsafe, beacuse the reducing of down speed started to late in the decend it did not really slow down at 0.5, no gentle landing this time.

Landing Protection (same for OA) is not 100% full proof. Sometimes i get in my drones a unsafe area this cx autolaning when i think  uh  why ?  looks safe to me  and v.v.


cheers
JJB


Landing protection

Sean-bumble-bee Posted at 4-11 13:11
OA off during landing because it can detect objects and prevent movement, which can cause problems. With the Mavic 2, DJI switch OA off automatically during automated landings, page 19 of the manual "Vision Systems are disabled during landing. Be sure to land the aircraft with caution.".
As far was I remember some of the fly app using drones have a similar line in their manuals.

Oke,

But if the vision system is disabled during landing....how can than the drone move away from an obstacle?
For this is needs the vision.

I do not have a Mavic2, nor the manual.Nothing in the MINI3PRO manual about disabling the vision during landing.
But in autolanding the OA is functioning (you will see the OA markings on the screen), but not used.
So you can land close behind a wall (few cm), drone will see the obstacle but will not move backwards.

For a MINI3PRO disabling Landing Protecion.


Landing Protection is active if OA is set to Bypass or Brake  and when applying down stick for landing.
If area in not safe, cancel landing and stick down will land craft. From that moment Landing Protection is OFF.

cheers
JJB

JJB* Posted at 4-11 23:26
Hi KingCon,

a) You saw your craft landing to fast....and bc of this touch down moved forward.

at the end the descend was reduced  but as said  little too late.
I don't know why I'm bothering to ask someone who doesn't answer questions, but anyway ....
The data shows that the autolanding slowed to <0.3 metres/sec before touching down.
That seems pretty slow.
How much should it slow to?

JJB* Posted at 4-11 23:41
Oke,

But if the vision system is disabled during landing....how can than the drone move away from an obstacle?

The M2P doesn't move away from objects during an automated descent, I presume the assumption is that an automated descent is only commanded when the drone is over the home point with a descent path that is clear obstacles or over a place that the pilot thinks is clear of such obstacles.
If OA is on during the Mini 3's automated descents then perhaps DJI changed the programming but I can imagine it making fine tuning of the descent path quite difficult if it prevents the drone moving horizontally towards an object and in that case I am sort of surprised that the Mini 3 does not have precision landing. Flying from around my house, especially from the back of the house, requires a vertical ascent/descent since the house and trees effectively ring the area. Imagine the OP's back garden paved area ringed by 60ft trees.
Without precision landing I am fair sure that the necessary small adjustments to the horizontal position of the Mini 3 would, in an automated descent with OA on, be trouble some.

Labroides Posted at 4-12 00:21
at the end the descend was reduced  but as said  little too late.
I don't know why I'm bothering to ask someone who doesn't answer questions, but anyway ....
The data shows that the autolanding slowed to

Don`t bother than, that is an easy fix for you.
Do you really think other forum users bother about you on here? they start to ignore you.

At least you do not call me an idiot.

If you, all on here, interested in a normal autolanding for a MINI3PRO with rate of descend values etc.
Excel file with data, make it easy to understand why the remote pilot in post #1 saw a too fast landing.
See that at touch down vertical speed is zero.
download link > https://mega.nz/file/YDRmDCSa#h9 ... WZNhzb3eDn-NZCiy774


cheers
JJB

Labroides Posted at 4-11 06:43
You really are the dumbest person in this forum.
You couldn't understand a single thing I've written here.
There are two many multi-syllable words for you to deal with.

What is your problem "cleaner wrasses"???

Why do you constantly need to insult people to pass your point across??

And where are the moderators for the DJI forum??? Can we stop this nonsense???

I mean, it clearly state in the Forum rules that:

II: General Forum Guidelines

5. We  will not tolerate rudeness, profanity (or attempts to bypass profanity  filters), insulting posts, personal attacks or purposeless inflammatory  posts.

Are this rules actually enforced or are they just for fun??? Can an Admin please explain why we need to tollerate this kind of constant crap???

Burt37 Posted at 4-12 01:40
What is your problem "cleaner wrasses"???

Why do you constantly need to insult people to pass your point across??

Totally agree,

If we all report this behaviour, than hopefully DJI will act on its own forum rules and simply BAN such people.

If your mindset is "i know everything" and by default all other are, let me prhase it friendly,  "not clever" , than you get such behaviour.
In psychiatrie they have names for that.

cheers
JJB

JJB* Posted at 4-12 02:34
Totally agree,

If we all report this behaviour, than hopefully DJI will act on its own forum rules and simply BAN such people.

I think a lot of people here are tired of this...

I'm just trying to spend some quality time reading some interesting fact about the DJI drones...

I'm a bit over, all this kindergarten attitude... Seriously, how old you need to be, to realise that that kind of talk is so childish and immature???

But, more importantly, why the administrators and moderators allow it to happen???

JJB* Posted at 4-12 01:16
Don`t bother than, that is an easy fix for you.
Do you really think other forum users bother about you on here? they start to ignore you.

Surprise surprise, yet another time you cannot directly answer a simple question and make smug dismissive remarks instead, just as you have for days on this incident.
But there's very little difference in the numbers, not enough to make any difference.

I used to think you could read and analyse flight data.

Labroides Posted at 4-12 06:00
Surprise surprise, yet another time you cannot directly answer a simple question and make smug dismissive remarks instead, just as you have for days on this incident.
But there's very little difference in the numbers, not enough to make any difference.

由软件自动封锁
Yóu ruǎnjiàn zìdòng fēngsuǒ

Labroides Posted at 4-12 06:00
Surprise surprise, yet another time you cannot directly answer a simple question and make smug dismissive remarks instead, just as you have for days on this incident.
But there's very little difference in the numbers, not enough to make any difference.

make smug dismissive remarks instead

You're kidding, right? pot calling the kettle...

He has answered, we all understand whats happened, its only you thats arguing this out.

You said in your reply that "its a mystery" lol. Its not a mystery, it landed too hard and bounced forward, it could only have done that if it was coming down too fast or landed on something with some give such as the middle of a car roof without a sunroof or landing extensions with some spring in them.

Either way, the bounce and what's left in the speed of the props or the turbulence from the very close quarters to the sleeper wall/steps on 2 sides caused it to move forward after the initial bounce but it should never have bounced in the 1st place, not from a normal landing on a solid surface. I land my Mini 3 Pro on my decking all the time when testing at home and its never once bounced, it comes down, slows, down some more, then slows right down to a crawl for the actual landing.

This does not look like a typical Mini 3 Pro landing, I am fairly certain that it should have slowed down more than it actually did else it would never have bounced, also, that camera returning to 0 degrees, i think that that was a little too late.

Bashy Posted at 4-12 06:50
make smug dismissive remarks instead

You're kidding, right? pot calling the kettle...

Just ignore this member in total

he is not worthy of any attention, well just my opinion and  i know this from more members on this friendly forum.

Bashy Posted at 4-12 06:50
Either way, the bounce and what's left in the speed of the props or

Actually I think in this post you are possibly in error. I would be surprised if much bounce was due to elastic rebound (think bouncing ball) or residual speed/energy left in the props/motor combination. I wouldn't mind a tiny bet that the drone's control system belatedly had some sort of reaction and revved the drone's motors.
It was/is one reason I was hoping the DAT would be available.

Burt37 Posted at 4-12 02:57
I think a lot of people here are tired of this...

I'm just trying to spend some quality time reading some interesting fact about the DJI drones...

Thank you for posting this. Im new to the forum but Ive already felt the burn. I almost decided to just stay away. Ive been on several other RC flying forums over the years - RCGroups, Flying Giants, RCU, various heli forums, sailplane forums, etc for many many many years. Im used to the mods stepping in much sooner than this with at least a warning or two if not a short term ban. A permanent ban would be guaranteed if it continued.
Now that I know Im not the only one, Im going to start using that report button if I think one of his posts breaks forum rules.

Thanks!!

Larry3215 Posted at 4-12 12:33
Thank you for posting this. Im new to the forum but Ive already felt the burn. I almost decided to just stay away. Ive been on several other RC flying forums over the years - RCGroups, Flying Giants, RCU, various heli forums, sailplane forums, etc for many many many years. Im used to the mods stepping in much sooner than this with at least a warning or two if not a short term ban. A permanent ban would be guaranteed if it continued.
Now that I know Im not the only one, Im going to start using that report button if I think one of his posts breaks forum rules.

Hi Larry

unfortunately the report button doesn't work, and judging by some of the reply, I think there a quite few bot in this forum, that definitely don't help when people take things way, way to seriously...

Burt37 Posted at 4-12 13:31
Hi Larry

unfortunately the report button doesn't work, and judging by some of the reply, I think there a quite few bot in this forum, that definitely don't help when people take things way, way to seriously...

That explains why he is getting away with that kind of attitude. This makes me far lass likely to participate. I suspect it is affecting others in a similar manner. Very unfortunate.