Meaning of a negative step response with quaternion

2018-08-15 02:37:52

It's not technically robotics but:

I've been trying to reproduce in Simulink a spacecraft attitude simulation using quaternions, and the kinematics and dynamics seem to work fine, however I'm having a bit of trouble with the controller.

I followed the model give in the 7th chapter which seems to be some sort of a PD controller. The control equation I used is:

$q_e$ is the quaternion error,

$\omega_e$ is the rotation speed error

But my results seems to be off.

With : Initial quaternion and rotation speed are $q_i = [0;0;0;1]$ and $ \omega_i = [0;0;0]$

I give a desired reference of $q = [0;1;0;1]$ and $ \omega = [0;0;0]$.

I get the following response:

$q(1)$ and $q(3)$ are staying at zero as expected.

But :

$q(2)$ is going towards -1 instead of 1 (As far as I understand the sign ambiguity does not explain this since q(4) is staying around 1)

$q(4)$ is not maintaining at 1. (I am not sure if this is related to the fact that the controller is only a PD)


  • It looks like you are negating your reference inputs in the summing junctions.

    2018-08-15 03:32:23