Skip to content

Conversation

rsilverioo
Copy link
Contributor

@rsilverioo rsilverioo commented Sep 22, 2021

The "kld_err" is initialized twice, and with the wrong values.
I guess you wanted to initialize the "kld_err" parameter to 0.01 and "kld_z" to 0.99.

The "kld_err" is initialized twice, and with the wrong values. I guess you wanted to initialize the "kld_error" parameter to 0.01 and "kld_z" to 0.99.
@rsilverioo
Copy link
Contributor Author

In fact I don't even know how you performed the localization with amcl correctly using that configuration (kld_err=kld_z=0.99).
You had to have very good and precise odometry 😅

Copy link
Member

@fjp fjp left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for spotting this!

@fjp fjp merged commit 11ac235 into ros-mobile-robots:noetic-devel Sep 22, 2021
@fjp
Copy link
Member

fjp commented Sep 22, 2021

In fact I don't even know how you performed the localization with amcl correctly using that configuration (kld_err=kld_z=0.99).
You had to have very good and precise odometry sweat_smile

I think the min_particles=500 "helped" to hide this issue, which is M_X_min in the KLD sampling MCL algorithm:

image

The odometry sensors I use (encoders in the DG01D-E motors) are definitely not precise ;-)

I hope to find more time soon to improve odometry and work on #37

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants