Hi all,

I'm a new P4P owner (just over a week) and I've been having a great time with it so far.
Tonight I decided to read more about how the AC behaves in RTH mode. Reading the manual (page 17) I am really struggling to understand what the following actually means:   

(I can't copy/paste from the PDF, that's annoying)
“Aircraft automatically descends and  lands if RTH is triggered when the aircraft flies within a 20 meter (65  foot) radius of the Home Point. Aircraft will stop ascending and  immediately return to the Home Point if you move the throttle stick if  the aircraft reaches 20 meters (65 feet) altitudes or beyond during  Failsafe.”

The first sentence is fine, I can understand what it's going to do. But the second sentence (in bold) is very, very confusing. Is it saying that if I ascend higher than 20m after takeoff (without first moving away a bit) that the aircraft will decide to just land? I'd really like to know what that sentence actually means because it seems kind of important.

Cheers

PS, yes I have read other threads but it still doesn't make any sense to me

Go here:
http://forum.dji.com/forum.php?mod=viewthread&tid=104450

Read posts #6, 7, 8, 9

Genghis9 Posted at 2017-8-27 05:58
Go here:
http://forum.dji.com/forum.php?mod=viewthread&tid=104450

Ok, I think I understand now. At least I wasn't the only person confused.

So, in summary let's say we have a situation where low battery triggers the RTH. Is the following statement correct?

IF the aircraft is greater than 20m in height (based on home position), but below the RTH altitude then using the left stick stop it going any higher and the aircraft will  then continue the rest of the RTH procedure at the altitude where you used the left stick without further user intervention.

After reading the comments you directed me to I think that's it, but please correct me if I'm wrong.

Thank you
Craig

I guess what scared me was that the paragraph in question mixes "yeah I'm going to automatically land" with "ok, let's fly back home at this altitude" (if the left stick was used).

CraigR Posted at 2017-8-27 06:17
Ok, I think I understand now. At least I wasn't the only person confused.

So, in summary let's say we have a situation where low battery triggers the RTH. Is the following statement correct?

Yes, that's basically it.  

It will land, either within the 65-foot radius or just immediately RTH and then land.  Either way you need to be extra vigilant in situations like this.  Many think that RTH is fully autonomous, as you can see that is only partially true.  In many cases, you retain some control, if necessary, to allow you to ensure the bird does not land in a place it should not.  In this example, if you were to do this there is a chance that it could hit an obstacle because you interrupted the process.  From reading a lot of threads, I'm convinced that many problems relating to RTH are because of situations like this.

  

One last important note, obstacle avoidance is disabled when the UAV begins descent to land, that can become a compounding problem if you take control during the landing if you don’t clear your path.

Fly Safe, Have Fun

CraigR Posted at 2017-8-27 06:22
I guess what scared me was that the paragraph in question mixes "yeah I'm going to automatically land" with "ok, let's fly back home at this altitude" (if the left stick was used).

Well as noted that is technically correct, but it depends on where the craft is based on the RTH point i.e. within the 65 foot radius of Home Pt. or not...

The real issue is this manual is very poorly written in English.  Likely due to a poor translation from Chinese/Mandarin, who knows.  They lack a technical writer who can clean up the wording, grammar, and logic to ensure it says what it needs too the way it needs too.  The good news is you have this forum, and there are many helpful folks here that can help clear this stuff up, either because they have the technical knowledge or actual experience or both.  Some may be a bit direct &/or rough around the edges, you'll just have to power through that and look for the information you need.
Good Luck to You!

Genghis9 Posted at 2017-8-27 10:33
Well as noted that is technically correct, but it depends on where the craft is based on the RTH point i.e. within the 65 foot radius of Home Pt. or not...

The real issue is this manual is very poorly written in English.  Likely due to a poor translation from Chinese/Mandarin, who knows.  They lack a technical writer who can clean up the wording, grammar, and logic to ensure it says what it needs too the way it needs too.  The good news is you have this forum, and there are many helpful folks here that can help clear this stuff up, either because they have the technical knowledge or actual experience or both.  Some may be a bit direct &/or rough around the edges, you'll just have to power through that and look for the information you need.

Thank you for the information which in my opinion is essential and best to learn these forums through people like you than after a crash for sure!

I wish I'd read your most recent reply before I took the bird out today for some flight tests regarding RTH because now I wonder what the behaviour is if I abort the RTH just after landing/descent starts... will obstacle avoidance turn back on? I did notice today in my tests that I got a message saying that it had turned off obstacle avoidance but I don't recall it telling me if it came back on after I aborted that landing. Maybe the logs will help although I have questions about those as well

Well, more tests tomorrow I guess! I'd better start writing some pseudocode and/or flowcharts to properly document the actual behaviour heh.

Cheers

I have forwarded to our colleagues and hopefully they can revise it for easy understanding.

CraigR Posted at 2017-8-29 00:34
Thank you for the information which in my opinion is essential and best to learn these forums through people like you than after a crash for sure!

I wish I'd read your most recent reply before I took the bird out today for some flight tests regarding RTH because now I wonder what the behaviour is if I abort the RTH just after landing/descent starts... will obstacle avoidance turn back on? I did notice today in my tests that I got a message saying that it had turned off obstacle avoidance but I don't recall it telling me if it came back on after I aborted that landing. Maybe the logs will help although I have questions about those as well

CraigR
OK good, however, your questions are starting to get out of my experience & knowledge level.  I don't know if it turns back on (obstacle avoidance), my guess is no they don't, or at least I'd assume that for now and act accordingly.
There are plenty of threads here on the logs but to date I have found none that really fully explain logs, that is a question for the experts.  I sincerely doubt the system will ever tell you that it has turned back on a feature or that it was restored, I'm betting from a programming side messages in general are negative (warning) types.  Most positive messages are few and only for obvious things like the uav is ready to go/fly.  My guess only, it is to keep the amount of code to a minimum.

Good Luck to Ya!

DJI Susan Posted at 2017-8-29 02:00
I have forwarded to our colleagues and hopefully they can revise it for easy understanding.

Susan
Thank you, P4P manual v1.2 really needs to be edited to improve readability and to include more important and relevant information to the UAV and its subsystems.

DJI Susan Posted at 2017-8-29 02:00
I have forwarded to our colleagues and hopefully they can revise it for easy understanding.

Thank you very much Susan.

Perhaps ask them if a flowchart would be more suitable than written language. I don't know what others think about this but I think that most people can already understand how a flowchart works and if they don't it's easy to learn. If the flowchart is accurate then ambiguities due to translation or prose or whatever essentially become irrelevant (because all the information is in the flowchart rather than a paragraph of text that attempts to explain what the flowchart would explain anyway).

CraigR Posted at 2017-8-29 09:09
Thank you very much Susan.

Perhaps ask them if a flowchart would be more suitable than written language. I don't know what others think about this but I think that most people can already understand how a flowchart works and if they don't it's easy to learn. If the flowchart is accurate then ambiguities due to translation or prose or whatever essentially become irrelevant (because all the information is in the flowchart rather than a paragraph of text that attempts to explain what the flowchart would explain anyway).

Generally I agree with that suggestion, there are several instances that would lend itself to using flow charts.
However, I also think that some accompanying text is necessary too.  I for one want too much information than too little, I can sort out the excess far easier than trying to fill in the blanks by missing details.  To be clear I'm not asking for technical schematics (I'm sure some would) and I'm not asking for programing code, but I am asking for detailed explanations on all areas.

Genghis9 Posted at 2017-8-29 09:11
Generally I agree with that suggestion, there are several instances that would lend itself to using flow charts.
However, I also think that some accompanying text is necessary too.  I for one want too much information than too little, I can sort out the excess far easier than trying to fill in the blanks by missing details.  To be clear I'm not asking for technical schematics (I'm sure some would) and I'm not asking for programing code, but I am asking for detailed explanations on all areas.

I think we're asking the same things, Genghis9. Yes accompanying text is necessary as well (for several reasons... redundancy[1] is a good thing for example, and there are other reasons as well even on the flowchart itself). I, too, would prefer too much than too little or, even worse, ambiguous information. Redundancy by providing written text that accompanies the flowchart would in 99.9% of cases probably mitigate or eliminate ambiguities. I sure don't want, or need, the entire logic going on beneath the "level of abstraction" necessary to safely and confidentially operate the UAV (i.e. the flowchart should not include so many details as programming code might do -- although if written correctly the code is probably top-down in design anyway so the details of how things are actually done are are at lower levels anyway). I'm certainly not asking for source code

Ugh. I forgot to add the footnote [1] and by the time I edited and hit submit the edit time had expired. So here it is here:

[1] I don't mean redundancy in the negative sense that it's, by some people, sometimes viewed today; i.e. as something that can be omitted. I mean redundancy in the positive sense used in: a) effective communication, where the same thing might be said twice but in a slightly different way or medium; and b) redundant backup or failsafe applications. If I could think of a term that has less potentially negative interpretations then I would, but redundant is the exact word I want as given by the example of flowchart + text.

If you read through the Forum, you'll find many pilots who's P4P either went missing or crashed because the RTH technology failed. Be careful.

CraigR Posted at 2017-8-29 09:45
Ugh. I forgot to add the footnote [1] and by the time I edited and hit submit the edit time had expired. So here it is here:

[1] I don't mean redundancy in the negative sense that it's, by some people, sometimes viewed today; i.e. as something that can be omitted. I mean redundancy in the positive sense used in: a) effective communication, where the same thing might be said twice but in a slightly different way or medium; and b) redundant backup or failsafe applications. If I could think of a term that has less potentially negative interpretations then I would, but redundant is the exact word I want as given by the example of flowchart + text.

CraigR
Agreed, I concur with both your posts on this