HLRobot_gazebo

A simulation for QKM HL6-0900 6DOF robot based on gazebo

Demo

Demo Link

Install

sudo apt-get install ros-melodic-ros-control ros-melodic-ros-controllers ros-melodic-gazebo-ros ros-melodic-gazebo-ros-control ros-melodic-hector-gazebo-plugins 
sudo apt-get install ros-melodic-sound-play 
sudo apt-get install ros-melodic-joint-trajectory-controller
sudo apt-get install ros-melodic-moveit*
pip install scipy
cd catkin_ws/src
git clone https://github.com/liuxiao916/HLRobot_gazebo.git
cd catkin_ws/
catkin_make

Launch

Just launch a robot.

roslaunch hlrobot_gazebo hl_gazebo.launch

Launch robot to play music.

roslaunch hlrobot_gazebo bringup_music.launch

Launch robot to plan with moveit.

roslaunch hlrobot_gazebo bringup_moveit.launch

Launch robot to start hand-eye calibration.

roslaunch hlrobot_gazebo bringup_calibration.launch

File Structure

File	Instruction
config	config file of controller in Gazebo
cubicTrajectoryPlanning	Include Forward and Inverse kinematics source file and motion planning source file to generate the music book. The music book (note and PPB) and the location of the insturment in Joint space (q_down) is also in it
data	The sound file of roll call and the picture for readme
launch	Launch file for simulation
meshes	The models for robot manipulators
rviz	The config file of rviz
scripts	Ros Python files
udrf	The description file of the robot manipulators

Control

In order to verify the forward and inverse kinematics, I use position_controllers/JointGroupPositionController for controller and PositionJointInterface for each joint. It means for every transmission the input is position and output is also position.

You can simply use this command to specify the angle (radian) of every joint.

rostopic pub /HL_controller/command std_msgs/Float64MultiArray "layout:
  dim:
  - label: ''
    size: 0
    stride: 0
  data_offset: 0
data: [0.426645, 0.515256, 1.761281, -0.169471, 0.998398, 5.370273]" 

For Moveit, I change the controller into position_controllers/JointTrajectoryController.

Inverse Kinematice

Set IKsolver.cpp to solve the inverse kinematic and control the robot. You can also use moveit now.

Frame

Frame base_foorprint and tool_frame are used to calibrate the world frame.(To get the world frame in lab) If you want to know the coordination in world frame, you can check the transform between them.

rosrun tf tf_echo base_footprint tool_frame

OR

rosrun hlrobot_gazebo show_tf.py

Compare the frame in simulation and reality

Load music book and play music

Edit the path of music book(PPB files) in scripts/publisher.py
Music book is in cubicTrajectoryPlanning/data/PPB
Edit the path of the cubicTrajectoryPlanning/data/q_down.txt in scripts/player.py

rosrun sound_play soundplay_node.py
rosrun hlrobot_gazebo play.py
rosrun hlrobot_gazebo publisher.py 

OR

roslaunch hlrobot_gazebo bringup_music.launch
roslaunch hlrobot_gazebo play_music.launch

Hand eye calibration

Use this robot to get the data for hand eye calibration.

roslaunch hlrobot_gazebo bringup_calibration.launch

Use image_view to collect image.

rosrun image_view image_view image:=/camera/image_raw

Get pose

rosrun hlrobot_gazebo show_tf.py

Todo

• load angle of joints from txt file to control the robot
• Try to play music in gazebo
• Show world coordination
• Implement the forward and inverse kinematics in this simulation
• Simply control the robot by giving Cartesian coordinate
• Moveit!
• Hand eye calibration!
• Anything useful