paparazzi-devel
[Top][All Lists]
Advanced

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

Re: [Paparazzi-devel] Mediatek for rotorcraft ?


From: Christophe De Wagter
Subject: Re: [Paparazzi-devel] Mediatek for rotorcraft ?
Date: Sat, 7 Jul 2012 21:38:55 +0200

Dear Jacky,

I'm sure your MediaTek receivers DO give good position measurements. So they can be used. What Felix meant are 2 things: 

- paparazzi works in LTP (local tangent plane) and converting from NMEA LLH to LTP is quite a lot of work for the processor and causes extra delays. Probably for that reason nobody implemented it yet (I think). This does not mean it can not be done. It just means you might still have to do it.

- in control theory, it is not just the resolution or precision of your measurement that counts, but also (even: mainly) the delay or "phase". For controlling a quadrotor you need to cancel lateral accelerations (attitude trim). If you have to wait for an acceleration to become a velocity and then even further wait until this velocity grows big enough to become an actual displacement that is larger than your measurement quality, then you will have a system with "poor control performance," meaning you can expect it to need significantly more space to hover in case of perturbations. If your GPS has a good and fast velocity measurement, your control loops are 1 step ahead (so-called 90 degree of phase lead) as they can start correcting before any apparent position error occurs. Mediatek, from my understanding fails to report the required speed information. This will degrade performance. (Do not use hff=horizontal feed forward filters integrating accelerometers as these will become unstable without velocity measurements) All of this still does not mean it is impossible. If your platform is sufficiently stable, your perturbations sufficiently low and you have sufficient space (low need for control performance) then I bet they will do just fine. 

Just my 2cent,

-Christophe 



On Sat, Jul 7, 2012 at 4:15 PM, Jacky COUSSIRAT <address@hidden> wrote:

Hi Felix 

I will not insist, but I always try to understand

Recently I bought DIYDrones APM2 which is equipped of Mediatek 

(the same with which I want to equip my LISA/M v2.0  )

I made several tests

 

 

If you open attached files you can see this APM2 works very well

 

How do they do themselves ...?

 

Cheers

 

Jacky




> Message du 04/07/12 23:39
> De : "Felix Ruess"
> A : address@hidden
> Copie à :
> Objet : Re: [Paparazzi-devel] Mediatek for rotorcraft ?

>
>
Hi,

>
As is, you can't use the mtk gps for rotorcrafts because it is not reporting velocities and ECEF position.

>
Cheers, Felix
>
>
On Wed, Jul 4, 2012 at 6:17 PM, Coussirat Jacky <address@hidden> wrote:
>
Hello  Felix
> Sorry but I would like to understand
>
> http://paparazzi.enac.fr/wiki/Subsystems
>
>
NameTypesFirmwaresArchitectureDescription
gps
  • ublox
  • ublox_utm
  • nmea
  • mediatek_diy
  • skytraq
  • all
  • fixedwing
  • all
  • all
  • rotorcraft
  • all
GPS drivers

> http://paparazzi.enac.fr/wiki/Subsystem/gps
>
>
typearchitecturesfirmwaresnotes
ublox all all recommended
ublox_utm all fixedwing for the older ublox4 series that still have a UTM message
skytraq all rotorcraft some fields missing in the skytraq protocol for fixedwings, see #167
mediatek_diy all fixedwing
>
nmea stm32 all only basic position and fix information, see #120

> What should we believe
>
> Thanks for your response
> Jacky Coussirat
>
>

> _______________________________________________
> 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



reply via email to

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