====== Preface ======
This tutorial will teach you how to get the PR2 to rotate the object in front of one of its cameras in order to obtain the object model.

====== Steps to reproduce the experiments: ======
  * bring up the robot
  * install and compile //ias_drawer_executive package//, should be newer than feb 17, 2011 10am and come from the tumros-internal git.
  * roslaunch ias_drawer_executive wopcl.launch
  * position the robot in front of the neutral background (e.g. white wall) (joystick is on)
  * bring the hand in the position in front of the camera:

  bin/ias_drawer_executive -121 1 -0.026910495869662321, -0.052571707068107605, 1.30157109054388, -0.67400052002840583, -1.3878958917494466, -1.3635043565396017, 0 .01

  * note: you can also use different positions for the object, put the arm into a nice position with the gripper approximately upright, then run >rosun tf tf_echo base_link l_gripper_tool_frame, this gives you a coordinate. then you put this into rosrun ias_drawer_executive ias_drawer_executive -2 arm x y z 0.707 0 0.707 0. with arm=0 for the right or =1 for the left arm. This will bring the arm into that position, now with the gripper pointing straight up. You can then read the joint angles from rostopic echo -n 1 /l_arm_controller/state and put them int the -121 line from above, which basically is -121 for rotate endless, 1 for left arm, then the 7 joint angles and finally the rotation speed in approximately radians per 1/20th of a second.
  * point the head: you can point get again the coordinate of the hand with rosrun tf tf_echo base_link l_gripper_tool_frame and then point the head there, you will notice that the gripper does not end up in the middle of the image, so some values that could go well with the above joint angles: bin/ias_drawer_executive -22 base_link 0.691, -0.08, 1.218
  * open the gripper:

  bin/ias_drawer_executive -18 [0|1]   with 0 = right, 1 = left gripper
  
  * close the gripper gently:

  bin/ias_drawer_executive -30 [0|1]   with 0 = right, 1 = left gripper

  * squeeze the gripper tightly (here the left one=1):

  bin/ias_drawer_executive -38 shows the current closing of grippers in meters
  bin/ias_drawer_executive -19 1 amount_in_% (%0 - fully opened, 100% is fully closed)

  * record images and TF in a bag:

  rosbag record /narrow_stereo/left/image_rect /tf -O bagname 

  * run object modeler:

**TBD!**