search

PlayUAV / PlayuavOSD

A Graphical OSD for FPV
http://en.playuav.com
GNU General Public License v3.0
67 stars 60 forks source link

Home arrow issue on latest code #29

Open ABLomas opened 8 years ago

ABLomas commented 8 years ago

(note: i see similar issue here: http://en.playuav.com/question/6 - but maybe github issue tracker is preffered?) So, in short: latest PlayUAV OSD code, latest (stable) plane software on pixhawk (pix mounted with rotation and this is set in software - YAW_270). OSD home indicator points to different direction (in fact, i think it points into opposite direction, rear). Flight video (video is very, very bad quality, but this is DVR issue - it corrupts entire video frame when encounters even slight hsync loss, real in-flight video was good enough for entire flight): https://youtu.be/dk8vGTCZvl0 (turn down sound) Logfile from pix: http://goo.gl/KmXpJh Flight KMZ file: http://goo.gl/ENIj78 (for loading to google earth, flight track)

Now about specific times in video - i guess, it was displaying home in wrong location for all time, but i'm not sure that home is in opposite side. I think, it's "floating" around, not allways 180° direction: 8:26 - home displayed on right, but actual "home" is on left. Also, RTH engages in few seconds and steers plane into correct direction ("real home" direction, not "displayed" - opposite) From ~8:55 i'm flying away from home, according to OSD. But actually, i'm going sideways (distance from home barely changes, but with 70km/h+ it should increase very quickly, if i would be going away from home, as OSD shows) 10:11 again, distance to home DECREASES, but indicator shows that i'm going away from home ....and similar issues everywhere in late flight phase. But i think, it was displaying CORRECT direction in beginning of flight, and drifted away, For example, check 3:40 - RTH steers plane into correct direction and arrow points into "H" letter (well, almost, but there may be side wind and AP compensates for that)..

ABLomas commented 8 years ago

Also, i have few more logs, including:

TobiasBales commented 8 years ago

So I just ran through the calculations and I think the distance/speed dividers for imperial are mixed up but for metric it should be fine. I did the calculation how the gps distance is calculated and it was fine (within 150m at ~50km range) so I have no instant idea what the issue might be.

The video isn't available anymore, would you mind uploading it somewhere and pass me the accompanying logs so I can take a look? Would be awesome!

ABLomas commented 8 years ago

My earlier video was deleted with YT account, so i just uploaded another example of bad distance calculation. My approximate route (ellipse, not on foot - only start-stop points marked) is here: https://goo.gl/maps/fKpK4pSEPYs KMZ file for this flight: http://lag.lt/screens/misc/lektuveliai/PPhantom/Logs/phantom_batt_cal_20160327.kmz Logs from pixhawk: http://lag.lt/screens/misc/lektuveliai/PPhantom/Logs/Phantom_batt_cal_20160327_home_glitch.zip Youtube video from DVR (too long to watch, see below): https://youtu.be/pw1FYE8kzk4 Now what's wrong with OSD: At 3:18 distance (bottom left, big letters) seems to be correct. I already see that wide route, distance is consistent with KMZ file or pix logs. Everything seems OK. Then distance/alt data is frozen up to 3:45 and boom! 3:47 distance suddenly becomes 4.04km. +700 meters out of nothing ;-) Distance again frozen up to 4:18 and again suddenly becomes 3.33km

Later calculations seems to be correct, but i'm not really sure - i can already see me clearly, but OSD still shows 1.2km distance. But this may be just optical glitch, i changed lens on my FPV cam...

This "wrong distance" thing is reproducible, i experience it in almost every flight, but in slightly different way (no "frozen calc", no jumps, but distance shown is completely unrealistic). GPS is fine, HDOP is < 1 over flight. I can do more videos if needed, this was just random DVR record i can find on my HDD, to replace initial question on 1st post.

TobiasBales commented 8 years ago

Thanks, I will take a look and see if I can find anything wrong. Have you tried a minimosd on the same platform and it does not exhibit the issues? (I would guess so, but just to make sure it is the osd in this case)

ABLomas commented 8 years ago

No, i did not tried MinimOSD on this platform. Also, logs shows correct distance, also no jumps on kmz file; RTL works correctly - so it's on OSD side, not FC...

TobiasBales commented 8 years ago

There is the telemetry component, but it is what I figured, just wanted to ask just in case ;) Which version of the osd firmware are you running? The last one downloaded by the windows config tool?

ABLomas commented 8 years ago

I'm using latest version, available from "official" flash tool.

macfly1202 commented 8 years ago

I've made lot of check long time ago and everything was ok with metric unit. I had ask Tom to add some relative Alt and other stuff and he do it without error. I've check with mission planner and nothing was wrong. I've got a custom antenna tracker (ASM), with arduino 2560 and distance was similar from ASM and playuav osd when I plug them to mavlink hub through mission planner output (ctrl+F mavlink serial output). I will check this again in few day but I take lot of time to check all distance and alt last year.

Envoyé de mon iPhone

Le 10 avr. 2016 à 14:52, ABLomas notifications@github.com a écrit :

I'm using latest version, available from "official" flash tool.

— You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub

macfly1202 commented 8 years ago

@ABLomas : what speed are you using for telemtry ?

ABLomas commented 8 years ago

Default - not changed SR parameters.

ThisIsTheOnlyUsernameAvailable commented 8 years ago

After resolving some unrelated issues on my 'copter, I took it for a relatively short flight to test capabilities.

I too experienced the home location being incorrect and actually got lost. Thankfully I was saved by RTH but experienced a serious 'brown pant' moment in the process.

I am using metric figures.

A serious issue that needs resolving, IMO.

TobiasBales commented 8 years ago

Yeah, I am trying to figure out what is wrong. A couple of people use it without having any issues and so far I was not able to find out, what is causing the issue.

I know it is probably not helpful but would one of you be willing to swap it out for a minimosd extra and see if the issue persist? (Just to rule out mavlink telemetry issues).

Also @ABLomas could you provide me with a log file + video where the gps does not lock up and maybe a rough idea on how to get them in sync ;)

ThisIsTheOnlyUsernameAvailable commented 8 years ago

Thanks Tobias.

Please let me know if you need any further detail.

ABLomas commented 8 years ago

I'm not really sure what you did mean with "gps not lock up"? I can do several flights and record jumping home position again ir required, but i usually keep GPS lock all time, so it should be not related to GPS. Anyway, looks like this issue affects several ppl, for example see last post: http://en.playuav.com/question/6

Having in mind that "radar display" is confusing itself - maybe just ditch it at all and just add simple "home arrow" (points to top => you fly home, points down - you fly away; in other words - just home direction relative to aircraft (airplane or copter))?

TobiasBales commented 8 years ago

The current version does have the simple home arrow and I am aware, that there are a few people having issues - trying to pin it down but it is not as easy and I haven't been able to fly/test myself in a while

IdefixRC commented 7 years ago

Hi guys,

Just found this one after I opened an issue on Tobi's github already. Same issue here. Botge radar and home arrow do not point in the right direction. Rather random actually :-(

trgFPV commented 7 years ago

The workaround at the moment is to set compass_use = 0 in your Ardupilit configuration. We are still working on this issue.

TobiasBales commented 7 years ago

Only do that on fixed wing though!

IdefixRC commented 7 years ago

Thanks guys

On Tue, Jul 12, 2016, 2:52 PM Tobias Bales notifications@github.com wrote:

Only do that on fixed wing though!

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/PlayUAV/PlayuavOSD/issues/29#issuecomment-231952757, or mute the thread https://github.com/notifications/unsubscribe/AI-c8PnmMw-01gvCvyY8r5PIp4ZLhIWMks5qUzmXgaJpZM4G7YIN .

mariansoban commented 7 years ago

hi there. anything new about this issue - especially for Copter?

ThisIsTheOnlyUsernameAvailable commented 7 years ago

There have been several code updates that seem to fix the issue. I recommend you follow the thread on DIYDrones which provides the latest information. Look at the last few pages of the following thread:

http://fpvlab.com/forums/showthread.php?40658-A-Graphics-OSD-for-APM-Pixhawk

StewLG commented 7 years ago

mariansoban:

DivotDigger has it (mostly) right*. Although the direction-to-home issue in PlayUAV has been the single most painful bug in PlayUAV by a very wide margin, five months since the last relevant patch to it we are starting to become confident we have actually fixed it. We have had no confirmed sightings recently that couldn't be pinned on running an older version.

You'll want to be running the so-called "Tobias Fork" of PlayUAV. You can download the Chrome configurator here:

https://chrome.google.com/webstore/detail/playuav-osd-configurator/clledgfbcikcmblfhbkhjeoebioekcnb

If you want some background about the fork, you can read it here:

http://en.playuav.com/question/43

For those who want to track the ongoing progress of the firmware, the best single place is the FPVLab forum referenced above (http://fpvlab.com/forums/showthread.php?40658-A-Graphics-OSD-for-APM-Pixhawk). We do try to follow all the various other places issues are posted as well, but sometimes things get past us -- like mariansoban's question.