Bitcraze Forums

Hello everyone,

I've been looking for a way to accomplish a "position hold" for the crazyflie when flying outdoors at considerable altitudes. GPS seems to be a good candidate for this application, and I read a couple old GPS-related forum posts that were very helpful, but before I got started with programming I wanted to ask a few questions to make sure that I wouldn't be "in over my head". First, the unit I'm considering using has a NMEA output at 1Hz, and I wanted to know would implementing a checksum and a string parser be too resource intensive for the crazyflie? In the forum posts it seems that uBlox was used due to it more resource efficient. Second, can the crazyflie magnetometer be used to obtain a magnetic north heading, or do the magnets in the motors, even when not running, skew the results enough to make it unreliable?

Thanks for any help or additional information!
Luke

Hi Luke,

There will be no problem with parsing NMEA and calculating the checksum, there is quite some processing power left in the Crazyflie. Using the binary protocol from ublock makes it much more efficient and easy to parse but we anyway came back to NMEA in our experiments since it allows to work with more gps. There is some NMEA parsing in the Crazyflie already: https://github.com/bitcraze/crazyflie-f ... rc/gtgps.c

As for the magnetometer there has been some work on calibrating it, after calibration the assumption is that it should be quite usable: viewtopic.php?f=11&t=2432&p=12375&hilit ... lib#p12375