Closed ahendrix closed 11 years ago
[watts] The 0.045 and 0.0305 might be offsets from the head plate frame.
[hsu] if that's the case, the offset should be moved to within the transform from head_plate_frame
to double_stereo links, i.e. line 60 or pr2_description/robots/pr2.urdf.xacro
. Offsetting between the camera frame and the camera optical frame does not seem correct.
[hsu] patch checked in r28462. Need some user testing.
[vpradeep] Looked at stereocam frames in rviz on a pr2 beta robot, and they seem correct, so closing ticket
urdf/sensors/stereo_camera.urdf.xacro line 61 contains hardcoded offsets: {{{
}}} specifically, the 0.045 and 0.0305 offsets are causing stereo point clouds to be "wrong" in simulation. Parameters
cal_stereo_*
appear to be zeros. (Note the rotation is needed here because optical frames are z-forward).trac data: