[Canberrauav] Flying tomorrow
christopher.d.gough at gmail.com
Sat Jan 14 23:18:05 EST 2012
You probably want to get to the bottom of the MTK GPS issues anyhow, but...
I can field two BirthdayGPS' (ublox neo-6Q) for the mugin plus keep a
spare handy (in my quad) in case it's required. One benefit of this is
that they can be simultaneously connected via UART and USB, so the
PandaBoard could access both of them directly without becoming
critical to either autopilot.
Or a GPS-management raspberry pi, if they ever become available :)
On 1/14/12, Andrew Tridgell <tridge at samba.org> wrote:
>> However, when you look at the xbee kmz file we made on the first flight of
>> the Mugin a few days ago, the xbee is solid, just like the Hope.
> As you have seen from my msg on the ArduPilot list, there is indeed more
> to this. The HopeRF and Xbee do actually produce very similar packet
> loss rates.
> The problem is that the MTK GPS attached to the APM1 on the Mugin lost
> lock during those 'missing' periods of the log. The same sort of MTK GPS
> attached to the APM2 didn't lose lock.
> I think that normally when a MTK GPS loses lock it keeps updating the
> lat and lon values with its best guess. The logs show a constant value
> during the periods where lock was lost. That makes me think that the
> serial stream between the APM1 and the GPS was interrupted, and perhaps
> the GPS itself was fine.
> If you can have a look at the cable between the APM1 and the GPS and see
> if there is any chance it could have been intermittent during the flight
> that would be a big help.
> Cheers, Tridge
> Canberrauav mailing list
> Canberrauav at canberrauav.com
More information about the Canberrauav