Add measurement update with elevation laser ranging in the EKF #174
Conversation
| @@ -1116,7 +1134,3 @@ static const DeckDriver vl53l0x_deck = { | |||
|
|
|||
| DECK_DRIVER(vl53l0x_deck); | |||
|
|
|||
| LOG_GROUP_START(range) | |||
| LOG_ADD(LOG_UINT16, range, &range_last) | |||
| LOG_GROUP_STOP(range) | |||
There was a problem hiding this comment.
We should not remove this kind of log variables that could be used by other tools.
|
Hi Marcus, thanks for the PR! |
|
Hello Arnaud! I didn't quite know if this was the way to push into the master branch, so do let me know if there is something that I should do differently. I have now revised the code, which now works well and have included more rigorous outlier detection and the laser angle of aperture.
This doesn't work for some reason, you have to manually uncomment the
Do you know why this doesn't work? Could it be a bug somewhere? |
A measurement noise model has been added, allowing for smooth transition of elevation measurements when moving from the feasible mesurement region (<1.5 m) to the region where only MOCCAP/LPS is active. The implementation has been tested with TDOA and the scalar update EKF
|
Merged, thanks! |
Added support for laser ranging in the EKF such that measurements are taken every 10 [ms] and ququed in the EKF, where they enter the filter by means of a scalar update. The standard deviation of the measurements are set low, and should be made a function of the distance - but this version of the code is stable and has been used to fly with the LPS system.
Edited with a second commit so as not to make pushing into the Kalman filter the default setting.