- Description
- Required packages - Kinetic Version
- Test Velocity Control with RVIZ
- Test Velocity Control with Gazebo and CAM (Gazebo plugin)
- Optional commands (Related to Test Velocity Control)
- Test Computer Vision Tools with Kinect
- How to use PC cam to track ar_tracker_alvar markers
- Sending commands through the action server
- Connecting with real UR5
- Contributors
This project is related to the application of the Adaptive Artificial Potential Field with ArUco Marker for grasping and collision avoidance using velocity control of ur_modern_driver.
Please check the paper for further info: [http://proceedings.science/p/111278]
- LaR Gazebo - Copyright (c) 2019, Erick Suzart Souza. All rights reserved.
- MoveIt! Kinetic
- MoveIt! Python
- Robotiq Gripper
- Universal Robot
- ur_modern_driver
- URSim running if you're using the simulated robot
Install any dependencies you might have missed by using this command in catkin_ws folder
$ rosdep install --from-paths src --ignore-src -r -y
RVIZ mode is used when the real robot is connected. In this case, Gazebo will not start. Wait for the message "You can start planning now!" to show before running the next command.
$ roslaunch custom_codes APF_project_rviz.launch
NOTE: Remember to start URSIM before launching ur5_ros_control Start the ur modern drive to connect with URSIM (3.9.1 version) - Remember to set DHCP and check the ip in the terminal. The IP is 127.0.1.1 as default.
$ roslaunch custom_codes ur5_ros_control.launch robot_ip:=127.0.1.1
If you don't have a webcam connected, launch this:
$ rosrun custom_codes tf_node.py
Start the command_vel node. The following command will turn on orientation control (if desired).
$ rosrun custom_codes command_vel_rviz.py --armarker --OriON
NOTE: Gazebo simulation is not working properly due to the lack of gravity compensation controller. Velocity controller cannot work well without compensation in Gazebo.
Gazebo simulation is used if the real robot is not available. If it is available, please use RVIZ and only RVIZ instead.
NOTE: The Gazebo environment is started in paused mode (to set initial joint angles). Remember to PLAY the simulation after gazebo starts.
$ roslaunch custom_codes APF_project_gazebo.launch
Spawn a sphere model to represent the goal
$ rosrun custom_codes spawn_model.py
While gravity compensation is not yet implement, the gravity is set to zero for the robot to reach the goal. The real UR5 has it implemented in its controller. In other words, it is not necessary to have this kind of control when the real robot is connected.
$ rosrun custom_codes change_gazebo_properties.py
Start rqt_image_view and select /ur5/camera1/image_raw topic to see the image being published
$ rqt_image_view
Start the ar_alvar_marker using the webcam topic published by the Gazebo plugin
$ roslaunch custom_codes gazebo_CAM_alvar.launch
Link the frame published by gazebo_CAM_alvar with the support of the D435 CAM. It is done in order to publish fresh tf data when the robot moves (it needs to be further investigated).
$ roslaunch custom_codes tf_transforms.launch gazebo:=true
Start the command_vel node in order to check if velocity control is working properly (check arguments). The following command will turn on orientation control (if desired).
$ rosrun custom_codes command_vel_gazebo.py --armarker --OriON --gazebo
If you want to test velocity control with a dynamic goal published by a node (without Kinect), first run this node before command_vel.py and then run command_vel.py with --dyntest argument.
$ rosrun custom_codes publish_dynamic_goal.py
If required, check if joint_group_vel_controller is running by calling:
$ rosservice call /controller_manager/list_controllers
Launch kinect driver using iai_kinect2 package.
Please follow its instructions for installation in its default repository [https://github.com/code-iai/iai_kinect2]
$ roslaunch custom_codes kinect2_bridge.launch depth_method:=opengl reg_method:=cpu
Launch ar_track_alvar
$ roslaunch ar_track_alvar Kinect2_alvar.launch
Load the Kinect2 TF Frame
$ roslaunch custom_codes tf_transforms.launch kinect2_test:=true
Remember to run command_vel node with --armarker argument
Calibrate your PC CAM using this package
http://wiki.ros.org/camera_calibration/Tutorials/MonocularCalibration
Install kinetic image pipeline
$ sudo apt-get install ros-kinetic-image-pipeline
Launch the TF_Broadcaster to transform the camera_link frame relative to the base_link of UR5.
If you don't have a web cam available use the tf_transform launch with argument fixed:=true instead.
$ roslaunch custom_codes tf_transforms.launch pc_camera_test:=true
Launch the PC CAM node to publish camera_link frame
Change video_stream_provider argument to '1' if an external USB CAM is used.
$ roslaunch custom_codes camera_pc.launch
Launch the ar_track_alvar node below Change the marker size (in centimeters) of the marker printed version in the launch file
$ roslaunch custom_codes PC_CAM_alvar.launch
If you want to test the position controller, sending commands directly to the /'controller_command'/command topic use
the following:
$ rostopic pub -1 /pos_based_pos_traj_controller/command trajectory_msgs/JointTrajectory "header:
seq: 0
stamp:
secs: 0
nsecs: 0
frame_id: ''
joint_names: ['shoulder_pan_joint', 'shoulder_lift_joint', 'elbow_joint', 'wrist_1_joint', 'wrist_2_joint', 'wrist_3_joint']
points:
- positions: [1.57, 0, 0, 0, 0, 0]
time_from_start: {secs: 1, nsecs: 0}"
Firstly check the machine IP. The IP configured on the robot must have the last digit different.
$ ifconfig
Disable firewall
$ sudo ufw disable
Set up a static IP on UR5 according to the following figure
Set up a connection on Ubuntu according to the following figure
Use the following command in order to connect with real UR5. If you are using velocity control, do not use bring_up. Use ur5_ros_control instead.
$ roslaunch ur_modern_driver ur5_ros_control.launch robot_ip:=192.168.131.12
This work was made possible by the teamwork of the LaR fellows:
- Caio Viturino - caiobarrosv
- Henrique Poleselo - hpoleselo
- Ubiratan de Melo - Ubira
Supervised by:
- Prof. Dr. André Gustavo Scolari Conceição - ascolari