AMCL ignores map orientation.

[dornhege]

Map Orientation is never used in amcl.

For orienation != 0 localization poses are wrong and initialization is likely to fail.

If not easily fixable at least users should be made aware of that when such a map is supplied.

Similar to #166.

[mkhansenbot]

Will this bug ever be fixed? I see no assignee or milestone for it.

[DLu]

There is not any active development on AMCL at this time. Patches are welcome.

[mkhansenbot]

David, are there any plans to fix this in ROS 2.0? Are you re-writing any of this anyway for ROS 2.0? If I was going to fix this, would you prefer it be in ROS or ROS 2.0?

[mikeferguson]

I imagine that an actual port of navigation to ROS 2.0 is at least 2-3 years away given how high we sit on the dependency stack.

I see no reason why this can't be fixed in ROS 1.0, as long as we fix both this and the static layer (so that if someone had a rotated map both nodes will properly rotate it and they won't even notice the change).

Pull requests for any bug fix are very appreciated -- our maintainers have time to review PRs and get them merged and tested, but generally not enough time to do much serious development these days.

[mkhansenbot]

Mike, thanks for the quick reply. I'm going to scope the work it would take to fix this in ROS 1.0 then. I just wanted to ask first so that I didn't fix something that would be obsolete soon anyway.

[mikeferguson]

Adding "help wanted" tag -- AFAIK there are no active developers with a plan to tackle this (would also need to tackle the costmap-related issues in #166). If someone is interested in tackling this, please speak up!