| Note: This tutorial assumes that you have completed the previous tutorials: Robots/REEM/Tutorials/Launching a REEM Gazebo simulation, Robots/REEM/Tutorials/play_motion. |
 Please ask about problems and questions regarding this tutorial on answers.ros.org. Don't forget to include in your question the link to this page, the versions of your OS & ROS, and also add appropriate tags. |
Use MoveIt! with REEM
Description: This tutorial will show you how to play with Move It ! with the REEM robotTutorial Level: BEGINNER
Next Tutorial: Robots/REEM/Tutorials/navigation
Launch MoveIt!
For a kinematic version (very fast) without controllers:
roslaunch reem_moveit_config demo.launch
For a version using the simulated model of REEM:
roslaunch reem_gazebo reem_empty_world.launch roslaunch reem_moveit_config moveit_rviz.launch config:=true
Test Rviz MoveIt! plugin
Go to MotionPlanning (top left) > Planning Request > Planning Group And choose right_arm_torso.
Drag the ball marker on the right arm to a desired pose, adjust orientation with the wheel-like markers. 
Go to the Planning tab (down left)
1) Click Update (on the Select Start State) to update the initial pose to the current robot pose.
2) Click on Plan to create a plan from current state to the marker pose. You will see an animation of the plan.
3) Click on Execute to send the plan to the controllers so the robot moves. The robot should move to the desired pose following the previewed plan. 
Further information can be found in the MoveIt! tutorials, in particular about the Rviz plugin.
Test MoveIt! via code
You can follow the official MoveIt! Move Group interface tutorials, for Python and for C++.
Also we offer some alternative simple Python examples:
Clone the reem_snippets package from our reem-utils github:
cd ~/reem-sim_ws/src git clone https://github.com/reem-utils/reem_snippets
The code is very self explanatory with comments all around it.
Now open moveit_pose_goal.py for a pose-based goal:
rosed reem_snippets moveit_pose_goal.py
Or moveit_joints_goal.py for a joint-based goal:
rosed reem_snippets moveit_joints_goal.py
For grabbing joint poses you can use joint_states_group_grabber.py
rosrun reem_snippets joint_states_group_grabber.py -i
Which gives an output like:
[INFO] [WallTime: 1389717404.043611] [1602.174000] Node initialized. Ready to grab joint states. Interactive mode! Write a group name (or short group name) and it's values will be printed. (Write exit to exit). ['all_joints', 'left_arm', 'right_arm', 'left_arm_torso', 'right_arm_torso', 'torso', 'head', 'right_hand', 'left_hand', 'right_hand_all', 'left_hand_all'] ['a', 'la', 'ra', 'lat', 'rat', 't', 'h', 'rh', 'lh', 'rha', 'lha'] > right_arm Name = Joint Value ================= arm_right_1_joint = 0.36602641942 arm_right_2_joint = 0.115875053619 arm_right_3_joint = -0.204846821507 arm_right_4_joint = 1.36099376372 arm_right_5_joint = 0.971732788767 arm_right_6_joint = 0.102786035467 arm_right_7_joint = -0.815733486405 right_arm = ['arm_right_1_joint', 'arm_right_2_joint', 'arm_right_3_joint', 'arm_right_4_joint', 'arm_right_5_joint', 'arm_right_6_joint', 'arm_right_7_joint'] [ 0.36602641942, 0.115875053619, -0.204846821507, 1.36099376372, 0.971732788767, 0.102786035467, -0.815733486405 ]
Which can be copy-pasted in a Python file as lists.
