Replies to This Discussion

Permalink Reply by Dr Mike Black on May 26, 2010 at 11:47am

oh forgot to mention.....the gps is at 5hz

Also the locosys is an odd bit of kit to program (IMHO) deosnt always obey its commands..

• ▶ Reply

Permalink Reply by Yves Gohy on May 26, 2010 at 9:36pm

Hello, I noticed that too with Ublox. I searched for the cause. Having done some research, I found that it was the control of the checksum, the problem.

I changed the system, but there was still much error.

It seems that this was due to the calculation load.

I also understood that the RX buffer memory is not enough.

I do not know if it helps, certainly not. But now I have no problem and I know not why.

• ▶ Reply

Permalink Reply by Dr Mike Black on May 27, 2010 at 12:13am

Hi Yves,

Thanks for jumping in,,,,,,

Do you know what changed, which revisions of code, that solved it?

Thanks for the heads up on the root...

perhaps one of our fearless leaders know what was changed and how?

its a bit frustrating as im dying to fly the IMU, but dont dare as it stands...More reading required...lol

regards.

Mike.

• ▶ Reply

Permalink Reply by Yves Gohy on May 27, 2010 at 8:29am

Alas, I have no idea what has changed. And yes, for my part I have not yet dared to fly with my Easystar, because whenever there is something wrong in the simulation.

But now it sounds good. Banzai......

• ▶ Reply

Permalink Reply by Dr Mike Black on May 27, 2010 at 11:06am

Thank Yves, think were stuggling with teh same problem.

Ok guys, really need some help here...

The problems are increasing not decreasing...

Day 11 of getting no where...

In the old days, back when men were men and UAV was somthing the doctor treated you for...

we used thermopiles and code 2.4..in those heady days, of PID loops and sunny afternoons, we have had many flights....aircraft flying about with full AP control, twin stars.....easystars......UAV drones....even ones i was not supposed to use....we revelled in the warmth of the love and trust that was Ardupilot with shield.......

We had AP at 8 - 10+ miles out! 2000 feet up, never missed a beat! (once we got all the PID`s and niggles sorted)

Today......things for me are different.....since re-entering the world and getting back to teh fun things in life....

This AP and IMU is now driving me to distraction....

Things not working i can live with inconsistent findings i cant....

So after another day of testing teh system, born out of frustration i built another AP adn shield..just incase over the time my 2 working AP`s had been struck with muppetry, or at least i had done summet to upset them...

reloaded the code, checked adn re-checked teh .h

Fired it up.....and exactly teh same problems.....

Added to the GPS lock dropping in teh IMU, we have some other wierd faults..

Apparently its not religated to just NMEA...as Yves was having the problem with the ublox....what changed? what revision of code was it fixed in...and why isnt mine as i re-downloaded again last night for todays testing just incase..

Setup AP for FBW and Auto..thought to try some waypoints..not to fly just to see the reaction of the AP..

After discovering that the config tool, crapped out with no internet (not been run for so long)..this caused a problem....so went home....made some missions for testing back at work...Went back to work!

Config tool & GE really deosnt like no internet...thank fully after being home i have the places needed cached..

so......

reprogrammed AP again with waypoints..switched over....GCS confirmed.....adn techinally i think it was trying to go there..

then....all of a sudden the aileron servos went from extent to extent...rolling and pitching the IMU didnt make it any better.....

FBW seamed good.....is there something i need to knwo about how teh AP tracks to target? this is obviously somthing that cant be put in teh manual...but what is normal behaviour? is there a comparison from airspeed to ground speed?

i know in 2.4 days, i had allot of problems with throttle not doing as it should..we workd through that and sorted it..

This time i have to blow down teh Pitot HARD to get it to pull back, last time it wouldnt throttle up..

have read through the .h about how th e throttle works....tried it from 2 to 70 in defines...not making discernable difference.,..

So in conclusion..

with a internittent fault with the aileron output...(yes i tried in on all 3 boards now) is thi snormal?

GPS dropping...

throttle not pulling back.....

its all starting to ware a little thin.....

in the last 11 days, i have spent the best part of 7days and nights....trying to get it to teh point where i dare fly it....


I could really use some enthusiasm guys and a spoonful of help.

I know this works....you guys know this work Doug came 2nd with it (congrats by the way!)..now i need it to work..........i believe in DIYd, the guys and the knowledgebase whole heartedly, even to teh point of selling thier gear over here in the UK.....

some help please...i swear im losing my marbles!!!!

regards,

Mike,.

• ▶ Reply

3D RoboticsPermalink Reply by Chris Anderson on May 27, 2010 at 12:16pm

Mike,

Sorry to hear about these problems! I'm in Europe right now and can't test with hardware, but everything was working fine last week...maybe we're introduced a bug in some recent rev? I think you're the only person who has tried an NMEA GPS, so it wouldn't surprise me if there is an issue there, but we've been using uBlox with no trouble so it's worrying to hear that there are issues there.

To help us debug this, please test a minimal setup: just the AP (no shield) and IMU w/GPS, with the config set to airstart=1 and the D6/D8 jumper cable installed.

Do you get GPS lock?

Does it respond correctly in Stabilize mode?

• ▶ Reply

Permalink Reply by Dr Mike Black on May 27, 2010 at 1:01pm

HI Chris,

Europe!..anywhere nice?

Thanks for coming back to me....

OK...NMEA i expected to have to work on this as you said no one has used it...

I think for the most part teh parser is pretty much spot on.

ok....tap into data from GPS...full flow 5hz 2 lines required others turned off GGA and RMC i think off the top of my head.

IMU gets lock, that never drops...solid blue all the time (once locked)

If you watch the data out the IMU. its constant, always outputing the attitude data, with every 5th or 6th line GPS data.

!!!LAT:52725226,LON:1066500,SPD:0.0008¼ëÁrâ)›
JŠšŠ‚¢šbIM鵚bŠâ«JŠb"*Õþƺ‚ª¢²bQYéÃr‚‚‚‚’ÒZêÂr‚’ÒOW:0,***
+++ASP:-1,THH:73,RLL:0,PCH:2,CRS:-3,IMU:253,ch3_in:1671,***
+++ASP:-1,THH:73,RLL:0,PCH:0,CRS:-3,IMU:253,ch3_in:1671,***
+++ASP:-1,THH:73,RLL:0,PCH:0,CRS:-3,IMU:253,ch3_inz1671,***
+++ASP:-1,THH:73,RLL:0,PCH:0,CRS:-3,IMU:253,ch3_in:1671,***
+++ASP:-1,THH:73,RLL:0,PCH:0,CRS:-3,IMU:253,ch3_in:1671,+**
+++ASP:-1,THH:73,RLL:0,PCH:0,CRS:-3,IMU:253,ch3_in:1671,***
!!!LAT:52725226,LON:1066498,SPD:0.0000rŠ¢b[¹^jšbY:±:1
+++ASP:-1,THH:73,RLL:0,PCH:0,CRS:-3,IMU:253,ch3_in:1671,***
+++ASP:-1,THH:73,RLL:0,PCH:0,CRS:-3,IMU:253,ch3_in:1671,***
+++ASP:-1,THH:73,RLL:0,PCH:0,CRS:-3,IMU:253,ch3_in:1671,***
+++ASP:-1,THH:73,RLL:0,PCH:0,CRS:-3,IMU:253,ch3_in:1671,***
+++ASP:-1,THH:73,RLL:0,PCH:0,CRS:-3,IMU:253,ch3ßin:1671,***
!!!LAT:52725226,LON:1066496,SPD:0.3,BEZ:181,WPN:1,DST:5870546,BTV:0.0p:i;-0.00¼³<ÊJ+++ASP:-1,THH:73,RLL:0,PCH:0,CRS:-3,IMU:253,ch3_in:1671,***

Please ignore the garbled characters as this was noise on the data stream (now sorted)

but you can see the pattern...5 lines of angular data, 1 or 2 lines of GPS and angles

when it drops out you get GPS in increasing spaced intervals until its gone..AP lock out (IMU still locked)

after 30-50 secs the GPS starts coming back in 1, then 2 ....the back to every 6 as per..this is with no intervention from me...just watching.....

I assume that the IMU will stay locked as its stream is constant from teh GPS and is a good clean stream..

Tomorrow, i will do the basic AP set up as suggested adn try airstart...make some copies of the data and post here....

the wierd thing is that Yves was having this same effect with the data stream out of the IMU in the ublox he is using....so maybe this isnt a pure NMEA problem....

Stabilize.....ok....this is strange (to me anyway) if switched to stab, i can move teh Tx stick to port adn it responds after a fashion, it looks like its trying to compensate for flying..If i move teh stick to starboard, it wont go past centre...(this might be expected?)

if you move teh IMU around it is def compensating for any out of level movements....i.e as you would expect thermos to move..and the resolution is very high....i can see that it can def hold it self level, even in a high wind...

if i switch to Auto..i can still move the surfaces very slightly with teh sticks....this to me is very unexpected....

anyway...thank you again....it was all starting to get to me.....

being wrong is fine..as we occasionally get to be right....but this was doing me in! :)

hope your trip to Europe is enjoyable!

regards,

Mike.

• ▶ Reply

3D RoboticsPermalink Reply by Chris Anderson on May 27, 2010 at 1:10pm

Mike, that's very helpful, thanks. I'll ping Doug Weibel and ask him to look at this, but a few quick reactions from me:

1) The ability to manually "nudge" the aircraft when it's in auto mode is a *feature*, not a bug ;-) It's described in the manual here.

2) I've noticed that thing about it not going past center on the starboard side, too. Not sure what's going on there...if I'm not mistaken, it only shows up when you have GPS lock, so this may be some some funky gain error we've put in.

• ▶ Reply

Permalink Reply by Dr Mike Black on May 27, 2010 at 2:48pm

Thanks Chris,

Nudge.....i did wonder what this meant in real terms...now i see the correlation...i had read it, but figured that this would show when flying, or at least become more apparent..lol...superb idea....i knwo in 2.4 i would switch sometimes quite quickly to do just that, more so when setting up gains...

2) glad its not just me...and back to what i said earlier...its trying to find out whats normal.. if this is experienced then not to worry...all systems have thier idiosynchratic behaviours....i just dont know 2.6`s and 1.7`s yet.....2.4`s. well versed and found workarounds for things as required...

thanks again..and looking forward to Dougs thoughts..

just for the record, nothing i have said is meant as a criticism, im just having a hard time trying to work out what to expect and what not...as im sure you and the core guys can attest, this is a vertical learning curve.

many thanks.

Mike.

The work done here is truly superb..

Permalink Reply by Dr Mike Black on May 29, 2010 at 1:23pm

Ok Guys,

Chris, went back to basics as you suggested, removed the shield, set to airstart...

all fired up, ran the system for about 4 hours....on the ground(well workshop anyway)

All seamed to function as specified.


GPS keeps dropping out as documented still...but not the end of the world...

Sp programmed AP to Stabilised and Auto

Popped 4 waypoints in..(if it flew would try it see what happens..)

Anyway....after figuring that staring at it any longer wouldnt solve anything. so with a 'Devil may care' attitude went out to test it..

Stabilize function....VERY GOOD...so good in fact when set, the aircraft sat stright and level..in fact it was so good that when i tried to stear itwith ailerons, it refused adn went stright and level...when i applied rudder it fought the rudder and continued on the ame course...

well...i was convinced that it wasnt going to try a death spiral on me so hit Auto....

it turned quite sharply onto the WP i set...off it went...next WP..turned....

Then i think the GPS must have dropped again (didnt hook up the xbees and GCS for this test) as it was confused, flew straight and level, but didnt know where to go....switched back to man..brought it back..back to Auto...it must have found GPS again..and off it went...managed 3 of the 4 WP before heading off towards the coast at great speed..

so i would say that this is 80% successfull.....

The throttle wont pull back at all....but i guess this is a fettling issue, probably the fastest we have traversed WP`s with AP....:)

So we need to find out what is happening with the GPS system....also as described its poss not just NMEA as Yves was having the same issue with ublox....anyone have any thoughts? is there a common part of the code that processes data?

Play some more on the throttle system

For the record arduIMU blows the Thermos away in accuracy and smoothness of flight...only seen this kind of control on VERY expensive Autopilots off the shelf..(this is also probably due to the excellent code, well done)

regards,

Mike.

• ▶ Reply

Permalink Reply by Dr Mike Black on May 31, 2010 at 3:00pm

OK Chaps...

been searching around the site..and found a comment by Jordi.

'NMEA at 5hz will overflow the buffer. Cause you some problems'

Could this be why GPS fades in and out, almost like clockwork?

i realise that a ublox, would prob solve this, but still waiting for delivery from Sparkfun (DHL screwed us Royally by not informing us of the delivery and sent it back to the USA only to have sparkfun send it out again ARGGHHH!!!...so were still waiting!)

Also there are some fast GPS`s coming available and NMEA if def in thier output datasheets..

just fudging in the dark here really...any GPS/Code gurus got any thoughts on this?

regards,

Mike.

• ▶ Reply

ModeratorPermalink Reply by Morli on May 31, 2010 at 4:07pm

Good work on testing , Doc.
Thanks for the feed back.I am sure this will help lots of people. cheers

• ▶ Reply

• ‹ Previous
• 1
• 2
• 3
• Next ›
• Page