Unstable equations?

[kramer65]

I’ve just posted this about half an hour ago, but now my question seems to be hidden/removed from the forum. I’m not sure why, so I hereby post it again.

I’ve purchased the BerryIMU and I’m trying to get some correct readings using the code from python-BerryIMU-gryo-accel-compass/berryIMU.py. When I tilt the BerryIMU 90 degrees to the right I would expect X to be around 90 and Y to be around 0. X is indeed around 90, but when approaching the 90X, the readings for Y suddenly go from around zero to about 100.

I asked a question about this behaviour on robotics.stackexchange, and the answer I got was that the problem was “estimating angles in regions of instability” in which they suggest to “modify unstable equations”.

I also asked about this on the BerryIMU github page (here) but I got no reply.

[Mark Williams]

Hi, sorry for the late reply.
I am not sure why you other post didn’t get saved.

You are correct about the instability of the adjacent angle when going beyond 90 degrees.

We are working on code to fix this. Both none AHRS and AHRS.

Try and look around the net for AHRS code for the LSM9DS0. There is plenty out there and AHRS will fix your problem.

Mark --OzzMaker.com --

[kramer65]

Hi Mark. Thanks for confirming this.

The main reason I choose the BerryIMU over other IMUs was because it came with Python example code. However, this flaw in the code practically makes the BerryIMU useless for me.

I’ve looked around for other AHRS code for the LSM9DS0, but I’m having a hard time using it, also because many of it is C/C++.

When do you expect the working code to be released?

[Author]

Posts