Mrinal, I did what you suggested, and made a new node containing only the distance test code, and, to my surprise, it worked. Doing some debugging, and finding differences between my two tests, I discovered that the error was a single line I added to my CMakeLists.txt file: set_target_properties(distance_test PROPERTIES COMPILE_FLAGS "${OPENRAVE_CXXFLAGS}") This was there to allow me to use OpenRAVE classes in my application, but it seems that this was what was causing the errors. My guess is, certain OpenRAVE headers configure bullet differently. Thanks, Matt Klingensmith -----Original Message----- From: Mrinal Kalakrishnan To: User discussions Sent: Mon, Nov 22, 2010 9:18 pm Subject: Re: [ros-users] PropagationDistanceField usage? Hi Matt, On Mon, Nov 22, 2010 at 4:00 PM, Matt Klingensmith wrote: > Unfortunately the code you gave me produces the same results in cturtle (a > diagonal line in the xz plane at a 45 degree angle). I thought that it might > have been an issue with tf, so I used static_transform_publisher to publish > the points to the origin at zero rotation, and still had the same results. I'm not sure what the issue is. I'm getting the expected results under cturtle as well. The following images show the iso-surface at different distances, for a line of points on the y axis. As expected, they are cylinders of growing diameter. http://www-clmc.usc.edu/~kalakris/ros/df1.png http://www-clmc.usc.edu/~kalakris/ros/df2.png http://www-clmc.usc.edu/~kalakris/ros/df3.png If you can send me a minimal node that I can compile that reproduces your problem, I'd be happy to look at it. - Mrinal _______________________________________________ ros-users mailing list ros-users@code.ros.org https://code.ros.org/mailman/listinfo/ros-users