[ros-users] New package: 6D localization in OctoMap

[Thomas Moulard]

On Fri, Oct 5, 2012 at 5:02 PM, Armin Hornung
<HornungA at informatik.uni-freiburg.de> wrote:
> Dear ROS-users,
>
> Following up on the interest in 6D localization after my ROSCon talk about
> "ROS for Humanoid Robots", there is now a new package
> "humanoid_localization" available in the humanoid_navigation stack:
> http://ros.org/wiki/humanoid_localization
>
> It enables Monte Carlo localization in a 3D OctoMap based on laser,
> odometry, and IMU to estimate the full 6D torso pose of a humanoid. With
> some modifications, the code may also be relevant e.g. for UAVs. Directs
> support for point clouds from a depth camera will appear soon, and the wiki
> page will also get some final documentation updates.
>
> Relevant publication:
> A. Hornung, K. M. Wurm, and M. Bennewitz. "Humanoid Robot Localization in
> Complex Indoor Environments", IROS 2010.
>
>
> As always, feedback and comments are welcome!

Awesome! Thanks Armin!

I am looking forward to see an implementation relying on point cloud
instead of laser scan.
That would be more suited to HRP-2 ;)

Just one note: I would not have called this "humanoid" as it seems
that the component is totally
generic and could be used in other context (i.e. drones, for instance).

On the opposite, one thing that would be interesting and humanoid
specifics would be to consider that at all point of time, the robot is
not falling, i.e. a planar contact with the floor occurs and prevent
some d.o.f. from drifting (z, rx and ry). I am not sure however, how
to integrate this with your current work...

Best,
-- 
Thomas Moulard