Object Localization using AprilTag ROS for Robotics

Built With

• Ubuntu 18.04
• ROS Melodic
• OpenCV

Getting Started

1. Install ROS USB driver

   1. Clone ROS USB camera driver

      mkdir -p catkin_ws/src && cd catkin_ws/src
      git clone https://github.com/bosch-ros-pkg/usb_cam.git

   2. Create USB camera directory

      cd ..
      catkin_make

   3. Set up ROS environment

      source ./devel/setup.bash  # need to run this command everytime a new terminal window is opened

   4. Build ennvironment

      cd src/usb_cam
      mkdir build && cd build
      cmake ..

   5. Check available cameras/webcams (make sure camera is connected)

      sudo apt-get install v4l-utils
      v4l2-ctl --list-devices
      
      # sample output
      # Integrated Camera (usb-0000:00:1a.0-1.6):
      #         /dev/video0

   6. Launch USB camera

      cd ..
      cd launch
      roslaunch usb_cam usb_cam-test.launch

2. Install AprilTag_ros driver

   1. Clone AprilTag repo

      cd ~/  # or choose other installation location
      git clone https://github.com/AprilRobotics/apriltag.git

   2. Install AprilTag

      cd apriltag
      mkdir build && cd build
      cmake ..
      sudo make install

   3. Clone AprilTag_ros repo and install

      cd ~/catkin_ws/src
      git clone https://github.com/AprilRobotics/apriltag_ros.git
      cd ..
      catkin_make

3. Camera Calibration

   1. Install ROS camera calibration package

      sudo apt-get install ros-melodic-camera-calibration

   2. Perform camera calibration

      roslaunch usb_cam usb_cam-test.launch  # turn on USB camera
      rosrun camera_calibration cameracalibrator.py --size 8x6 --square 0.024 image:=/usb_cam/image_raw camera:=/usb_cam

   The example above uses an 8x6 checkerboard with 24mm squares (Calibration uses the interior vertex points of the checkerboard, so an "9x7" board uses the interior vertex parameter "8x6" as in the example below).

   Download the official ROS 8x6 checkerboard here.

   Move checkerboard up/down/left/right until "CALIBRATE" button turns green, then click on "CAILIBRATE", wait for the system to complete calibration calculation, then click on "SAVE", then "COMMIT".

   When you click on the "Save" button after a succesfull calibration, the data (calibration data and images used for calibration) will be written to /tmp/calibrationdata.tar.gz.

Usage

1. Using AprilTag_Ros and Detecting AprilTags

   1. go to /apriltag_ros/apriltag_ros/config and configure settings.yaml and tags.yaml.

      check out http://wiki.ros.org/apriltag_ros/Tutorials/Detection%20in%20a%20video%20stream for parameter setup.

      1. settings.yaml

         tag_family:        'tag36h11'
         tag_border:        1 
         tag_threads:       2 
         tag_decimate:      1.0 
         tag_blur:          0.0 
         tag_refine_edges:  1 
         tag_refine_decode: 0 
         tag_refine_pose:   0 
         tag_debug:         0 
         publish_tf:        true

      2. tags.yaml (Values based on the diagram below)

         standalone_tags:
           [
             {id: 10, size: 0.15},
             {id: 20, size: 0.1},
             {id: 30, size: 0.07}
           ]
         tag_bundles:
           [
             {
               name: 'my_bundle',
               layout:
                 [
                   {id: 0, size: 0.05, x: 0.0000, y: 0.0000, z: 0.0, qw: 1.0, qx: 0.0, qy: 0.0, qz: 0.0},
                   {id: 4, size: 0.05, x: 0.0548, y: -0.0522, z: 0.0, qw: 1.0, qx: 0.0, qy: 0.0, qz: 0.0},
                   {id: 3, size: 0.05, x: -0.0580, y: -0.0553, z: 0.0, qw: 1.0, qx: 0.0, qy: 0.0, qz: 0.0},
                   {id: 2, size: 0.05, x: 0.0543, y: 0.0603, z: 0.0, qw: 1.0, qx: 0.0, qy: 0.0, qz: 0.0},
                   {id: 1, size: 0.05, x: -0.0582, y: 0.0573, z: 0.0, qw: 1.0, qx: 0.0, qy: 0.0, qz: 0.0}
                 ]
             }
           ]

         
      3. Go to apriltag_ros/apriltag_ros/launch, add the following lines to continuous_detection.launch

          <arg name="camera_name" default="/usb_cam" />
          <arg name="camera_frame" default="usb_cam" />  
          <arg name="image_topic" default="image_raw" />

Usage

1. Launch camera

   roslaunch usb_cam usb_cam-test.launch

2. Launch AprilTag detection

   roslaunch apriltag_ros continuous_detection.launch

3. View results

   rostopic echo /tag_detections

4. Visualize results

   rqt_image_view

   select tag_detections_image from drop-down menu in rqt ImageView
5. Run RViz

   rosrun rviz rviz

   1. Select appropiate Fixed Frame.
   2. Click "Add" on bottom right in RViz, under "By display type", scroll down to "TF", select and click OK.
   3. The following interface should appear:

Detecting distance between two objects (tags)

See the open issues for a full list of proposed features (and known issues).

(back to top)