The conjugation for the qa

[dydyddy]

Hi, friend. We appreciate the code you uploaded.
But when I test the program, I have a problem to ask you.
In demo2, the converting of the quaternion to Euclidean space is

waypoint = (des_quat*qa).log()

Why is not

waypoint = (des_quat*(~qa)).log()

[lbusellato]

You are correct, it should be the quaternion conjugate.

On another topic, after some discussions with the author of the method, it seems that my implementation has some issues, especially regarding covariance estimation. I'm in the process of fixing them, but in another project, so I don't expect to be able to fix this repo before august. Also, I'm not even sure if those issues are present here, since I implemented KMP from scratch in that other project.

[dydyddy]

Thank you very much.

[kir0ul]

On another topic, after some discussions with the author of the method, it seems that my implementation has some issues, especially regarding covariance estimation. I'm in the process of fixing them, but in another project, so I don't expect to be able to fix this repo before august. Also, I'm not even sure if those issues are present here, since I implemented KMP from scratch in that other project.

Is it an open source project? If yes, may I ask which project it is?

[lbusellato]

Is it an open source project? If yes, may I ask which project it is?

Here. I am no longer actively maintaining it, so issues might take a while.