Improved gyro initialisation#456
Conversation
|
@digitalentity , this is some further cleanup of the gyro initialisation which brings it into line with some changes I made in betaflight a while ago. It doesn't directly address the memory problem, but I'd be interested to know if that problem still exists after these changes. |
| @@ -42,6 +42,7 @@ | |||
| acc_t acc; // acc access functions | |||
| int32_t accADC[XYZ_AXIS_COUNT]; | |||
| sensor_align_e accAlign = 0; | |||
| const uint32_t accTargetLooptime = 1000; | |||
There was a problem hiding this comment.
we can't assume acc is sampled at 1khz rate. it's sampled at the same rate as gyro at the moment.
martinbudden
force-pushed
the
inav_gyro_init
branch
from
4bde045 to
0a519ac
Compare
|
I tried this on top of master. It was unflyable with massive vibrations on 'roll' --- elvis on steroids. |
martinbudden
force-pushed
the
inav_gyro_init
branch
from
216328b to
61246b1
Compare
|
@stronnag , can you try this again when you get a chance? There was a silly mistake in the filter initialisation which I have now fixed. |
martinbudden
force-pushed
the
inav_gyro_init
branch
from
6a9d6a6 to
1ba65de
Compare
|
Tested. Works, no unexpected behaviours. Has no effect on #431. |
|
@martinbudden can you address the merge conflict? |
martinbudden
force-pushed
the
inav_gyro_init
branch
from
1ba65de to
b969f6f
Compare
|
@digitalentity , this is rebased and ready to merge now |
3 participants
No description provided.