This has been bugging me for a while now, and it's starting to make things quite difficult.
The point cloud data (and, if I remember correctly, stuff like odometry data) from the Turtlebot come in with 'depth' oriented along the z-axis, while 'depth' relative to the Turtlebot lies in the x/y-plane.
How do I get this to display and work properly together, especially in rviz?
Currently in rviz, the point cloud emanates from the 'top' of the Turtlebot, and points up in the air.
I'm sure there's got to be some sort of quick solution to this, so I figured I'd better ask before I go writing all of my own nodes and stuff to try to compensate for it.
My best guess is that there's some simple setup stuff I'm not doing properly or whatever. If there isn't actually a 'solution' to this because it's not a problem for you --if it 'just works' for you-- would you kindly walk me through how you set up/run your Turtlebot? I'm pretty sure I'm following the Turtlebot bringup tutorials and everything properly... but they don't talk about making it work with rviz, from what I can tell.
Thanks in advance!
Post score: 1
Comment by mmwise on 2012-01-27:
When the TurtleBot runs out of the box in rviz you should see the kinect pointcloud in front of the TurtleBot as shown in this video http://www.ros.org/wiki/turtlebot/Tutorials/Looking%20at%20Camera%20Data . It might be that your fixed frame is messed up somehow is it set to base_link?
Comment by patrick_hammer on 2012-02-19:
Works out of the box for me - does your robot model look correct in rviz?