Thread Tools
Nov 18, 2019, 02:08 AM
OlliW
Thread OP
the considerations are essentially correct, but you mix a bit yaw drift and horizon tilt. The g effect you describe results in horizon tilt, while the yaw drift is because the acceleration is a vector and not a matrix, so not g related
the differential measurement won't help, both are subject to the same issues
what helps is to seek info from the flight controller, if it has a full-fledged AHRS, which is what T-STorM32 can do (I call it then STorM32 Link)
Sign up now
to remove ads between posts
Nov 18, 2019, 05:15 AM
Registered User
Hi,

Quote:
you mix a bit yaw drift and horizon tilt.
I don't know what is the confusion, in a stabilization gimbal yaw drift and horizon tilt are correlated so...

Quote:
The g effect you describe results in horizon tilt, while the yaw drift is because the acceleration is a vector and not a matrix, so not g related
I agree with the first part of your sentence. I mean, the g effect of any moving vehicule will impact the measurement of the G force and induce a drift that will be compensated with time since the algorythme can not know in which direction the vehicule accelerates...

So this issue can only be worked around using a 2nd IMU on the vehicule itself. So I was wondering if in the case of the tests done by your first DIY tester the drift issue was resolved if he was using a 2nd IMU ?

I have another question about the magnetic position sensors that you use on each motor. Do the give you a very precise position of each motor ? At which rate do you measure them ? What is the big step of using them instead of integrating the gyros to get the position ? (more practical to use of more filtering ?)

Thank you for your replies, this type of project is complex but this is a lot of fun to understand the principles and limits of the sensors and the way to use the data they provide !

Regards,
Nov 18, 2019, 06:16 AM
OlliW
Thread OP
Quote:
Originally Posted by aerial_mars
I don't know what is the confusion, in a stabilization gimbal yaw drift and horizon tilt are correlated so...
LOL
and I was trying to indicate to you that you are mistaken

Quote:
Originally Posted by aerial_mars
So this issue can only be worked around using a 2nd IMU on the vehicule itself. So I was wondering if in the case of the tests done by your first DIY tester the drift issue was resolved if he was using a 2nd IMU?
I repeat my answer from before

Quote:
Originally Posted by aerial_mars
I have another question about the magnetic position sensors that you use on each motor. Do the give you a very precise position of each motor ?
I have no idea what you call precise
the TLE5012B are 15-bit sensors

Quote:
Originally Posted by aerial_mars
At which rate do you measure them ?
too slow, I need to improve that
currently at 1 kHz

Quote:
Originally Posted by aerial_mars
What is the big step of using them instead of integrating the gyros to get the position ? (more practical to use of more filtering ?)
I've discussed that at length in post #3, I'd like to refer to this
Nov 19, 2019, 11:13 AM
Registered User
Sorry I was thinking ROLL instead of Yaw of course ! That's why for me roll drift and level issue are correlated (It is a traditionnal issue on gimbals). I don't know why I made this confusion, it was late at night when I sent you my first message... lol

15 bit is very precise ! Cool...

Thanks for your reply, I will have a look at post#3 to read it again !


Quick Reply
Message:

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Cool T-STorM32: The best STorM32 ever !? OlliW Multirotor Drone Talk 2 Feb 21, 2017 05:02 PM