paparazzi-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Paparazzi-devel] NavSetGroundReferenceHere(), USE_INS_NAV_INIT and


From: Luke Ionno
Subject: Re: [Paparazzi-devel] NavSetGroundReferenceHere(), USE_INS_NAV_INIT and Reported Alt
Date: Fri, 7 Mar 2014 10:48:07 -0500

Sorry for the delay, things got a bit busy, but I’m back:

 

Paparazzi version: 5.0.3

LISA/M 2.0 airframe config:  (see attached)

LISA/M 1.0 airframe config:  (see attached – note that it’s using the 2.0 board definition, with the ADC scaling factor redefined.)

 

Physically, the LISAs are mounted on the bottom of the main frame of the quads.  The barometers are shielded from direct light. 

 

On a related note, I was wondering if it’s normal to see 2-3m drift in the altitude estimate, using a barometer and the rotary-wing firmware?  I’m flying three ships indoors (in an inflatable structure, so I still get good GPS) and I’m having trouble with the altitude estimate drift being about the same as my desired flight plan altitude (2-4m above the floor).  I just took some raw barometer readings in manual flight with a 2m tether, to make sure I held the same altitude, and I’m seeing a good deal of drift/variability (see attached), though admittedly, the errors are small, relative to the full-scale reading.  Again, if this is normal, I’ll just have to live with it (maybe install sonar), but I want to make sure, before I move on to plan-Bs.

 

Thanks again!

 

-Luke

 

From: address@hidden [mailto:address@hidden On Behalf Of Felix Ruess
Sent: Sunday, February 16, 2014 1:13 PM
To: Paparazzi UAV devel list
Subject: Re: [Paparazzi-devel] NavSetGroundReferenceHere(), USE_INS_NAV_INIT and Reported Alt

 

Could you please provide some more details, about your paparazzi_version, airframe config ,etc...

 

On Sat, Feb 15, 2014 at 11:14 PM, Luke Ionno <address@hidden> wrote:

Oh, my bad,  no, I’m not using the Baro_BMP module.  (I’ve got the autopilots installed in quads.)

 

-Luke

 

From: paparazzi-devel-bounces+nsknews=address@hidden [mailto:paparazzi-devel-bounces+nsknews=address@hidden] On Behalf Of Felix Ruess
Sent: Saturday, February 15, 2014 4:47 PM
To: Paparazzi UAV devel list
Subject: Re: [Paparazzi-devel] NavSetGroundReferenceHere(), USE_INS_NAV_INIT and Reported Alt

 

Regarding the baro on Lisa/M 1.0, just fixed a bug in in the baro_bmp module with commit d404719.

So if you were using that, could you please re-test?

 

On Sat, Feb 15, 2014 at 10:29 PM, Gautier Hattenberger <address@hidden> wrote:

Hi,

You can try to use NavSetAltitudeReferenceHere (instead of NavSetGroundReferenceHere), it use to work some years ago...
Question 1 may take a bit more time to solve

Gautier

Le 15/02/2014 21:54, Luke Ionno a écrit :

Hi all!

 

I’ve been experimenting with a couple Lisa/M 2.0s, and a Lisa/M 1.0, with/without USE_INS_NAV_INIT, and I’ve got a few couple questions:

 

With the two Lisa/M 2.0s, if I enable USE_INS_NAV_INIT, and don’t use the NavSetGroundReferenceHere function, they will start out reporting they’re at the flight-plan ground altitude (exactly what I’d expect), but then when they get a GPS lock, they jump to 80 meters, which is some 30 meters above the actual GPS height I get here (and 20 meters above the flight plan ground altitude.)  They do this consistently/precisely; on two different ships, across 4-5 boot cycles, I’ve been dead on 80m each time.  With or without GPS, the MS5611 barometers work as expected – height deltas are measured correctly.  If I use NavSetGroundReferenceHere(), I get the accurate/correct altitude of ~50m, albeit with the usual GPS variation. 

 

Now, If I use the exact same settings with the Lisa/M 1.0, the altitude starts out at around -500 to -600 meters, and stays there, either with or without GPS lock.  (As with the Lisa/M 2.0s, the BMP085 barometer works as expected.)  The only way to ‘fix’ the large negative altitude is to call the NavSetGroundReferenceHere() function, at which point, it starts reporting the accurate altitude of ~50 meters.  Note that the Lisa/M 1.0 isn’t 100% stock; I jumpered the USB-Detect pin, so that I could use the bootloader, and then used UART3 for the GPS, just like the Lisa/M 2.0s.  (It’s running the Lisa/M 2.0 firmware, in fact.)  I can’t see how this would be responsible for the absurdly negative altitude at boot, but I figure it’s worth mentioning.

 

So, question #1: Any idea why the 20m jump (to an inaccurate altitude) with the Lisa/M 2.0s, and any thoughts on what I could do to fix the large negative offset on the Lisa/M 1.0, so it at least does the same thing as the other two?

 

And, question #2: In general, I’ve been using NavSetGroundReferenceHere() when I fly, and that works fine for the most part, but there are times when I’d prefer it if my carefully-placed waypoints didn’t get moved around during Geo_Init.  Is there any way to ONLY initialize the height, so I get rid of the offsets described above, without moving the rest of the flight plan?  (Keep in mind that I’d be fine if it simply zeroed out the barometer/GPS altitude with respect to the flight plan ground altitude at boot; I’ve been flying some simple multi-ship formations, and it would be nice if they were all zeroed out the same, even if the height was not precisely accurate WRT MSL...)

 

Thanks!

 

Regards,

-Luke

 

 

 

 

_______________________________________________
Paparazzi-devel mailing list
 
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

 


_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

 


_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

 

Attachment: LISAM20.xml
Description: Text Data

Attachment: LISAM10.xml
Description: Text Data

Attachment: Raw_Baro.jpg
Description: JPEG image


reply via email to

[Prev in Thread] Current Thread [Next in Thread]