Thanks Doug! When these GPS units get here I'll give it a try, I agree with your comments about 10Hz and maybe I'll use the 5hz setting, there is also single sentence support that I will have a play with. It means you can customise the NMEA string to exactly what you want to make parsing easier although it is obviously not a standard NMEA 163 sentence.
David, I should have mentioned - if you want to try it out at 10 Hz, then plese look through the Ublox parser code. I have some performance reporting stuff in there. It currently is only there for ublox, but if you extended it to your case it will give some valuable feedback. In particular the metrics for maximum main loop execution time and # of main loop cycles in a reporting interval....
10Hz is probably no problem. The biggest issue would be handling the volume of associated serial I/O. If the gps has compact binary messages then it would probably be fine. 10Hz with big nmea messages to parse might pose an issue. Also, if you limit the volume of data that the IMU has to pass through from the gps to the AP....
As far as the DCM algorithm itself 10Hz would be no issue whatsoever. Also, I don't know how much advantage there would be to using a gps at 10Hz. Some, certainly, but would it be any significant improvement? Remember that all we are usign the GPS for w.r.t. the IMU is to get a speed value for centrepital correction and a ground course for yaw drift correction. The yaw drift correction has a time constant of a couple seconds, so 10Hz probably is of little advantage there. And most airframes will not have large speed changes over a short time period, so the advantage in centrepital correction is unclear.
Comments
Look for rev 49 under the source tab. It is easiest to retrieve current revisions if you use a SVN client.
Where are te code for EM406 and ArduIMU?
Thank You
10Hz is probably no problem. The biggest issue would be handling the volume of associated serial I/O. If the gps has compact binary messages then it would probably be fine. 10Hz with big nmea messages to parse might pose an issue. Also, if you limit the volume of data that the IMU has to pass through from the gps to the AP....
As far as the DCM algorithm itself 10Hz would be no issue whatsoever. Also, I don't know how much advantage there would be to using a gps at 10Hz. Some, certainly, but would it be any significant improvement? Remember that all we are usign the GPS for w.r.t. the IMU is to get a speed value for centrepital correction and a ground course for yaw drift correction. The yaw drift correction has a time constant of a couple seconds, so 10Hz probably is of little advantage there. And most airframes will not have large speed changes over a short time period, so the advantage in centrepital correction is unclear.