Peter: In general we don’t want to support fully generic AP_Periph builds. Andrew: Andy, are you trying to save CPU with this change?
Let’s try to see if there’s an automatic method to detect whether there has been any valid output command. If not, then we can go with this PR. P: We’re probably going to have issues with NeoPixels.
UTC0719
A: SITL testing for this would be fine.
The test could fly a hard bank in FBWA, then get the trigger for this PR and check if nav roll and pitch have updeated.
UTC0725
Merged!
UTC0726
Andy: Scripting has only access to the lastly calculated check in the main code. A: So we get a relative position as a result. But where is this relative to?
If we had access to the absolute position that this vector refers to, then this would be usable by the script to calculate the current location of the fence.
Why not return an absolute breach location?
Andy: Okay, I’ll adapt the API.
UTC0745
A: This feature depends on the define HAL_PROGRAM_SIZE_LIMIT_KB. If STRacing can’t get this feature, then that’s fine by me.
UTC0748
P: Question re. Math.h and Vector3.h.
They both have a (circular?) definition of is_zero() and could be untangled.