ROS 2 OpenVINO™ Toolkit Sample Application#

This tutorial tells you how to run the segmentation demo application on both a static image and on a video stream received from a Intel® RealSense™ camera.

Run the Sample Application#

  1. Check if your installation has the amr-ros2-openvino Docker* image.

    docker images |grep amr-ros2-openvino
#if you have it installed, the result is:
amr-ros2-openvino

    Note

    If the image is not installed, continuing with these steps triggers a build that takes longer than an hour (sometimes, a lot longer depending on the system resources and internet connection).

  2. If the image is not installed, Intel® recommends installing the Robot Base Kit or Robot Complete Kit with the Get Started Guide for Robots.

  3. Check that EI for AMR environment is set:

    echo $AMR_TUTORIALS
# should output the path to EI for AMR tutorials
/home/user/edge_insights_for_amr/Edge_Insights_for_Autonomous_Mobile_Robots_2023.1/AMR_containers/01_docker_sdk_env/docker_compose/05_tutorials

    If nothing is output, refer to Get Started Guide for Robots Step 5 for information on how to configure the environment.

  4. Launch the automated execution of the ROS 2 OpenVINO™ toolkit sample applications:

    docker compose -f $AMR_TUTORIALS/ros2_openvino.tutorial.yml up

    Expected output:

    1. Execution of the object segmentation sample code input from the image: This takes one minute, and you can see the semantic segmentation being applied to the image.

      Original image

      ../_images/expressway_openvino_orig.png

      Image with semantic object segmentation

      ../_images/expressway_openvino_with_segmentation.png

    2. Execution of the object segmentation sample code input from the Intel® RealSense™ camera topic: This requires a Intel® RealSense™ camera connected to the testing target. It takes one minute, and you can see the semantic segmentation being applied to the video stream received from a Intel® RealSense™ camera.

  5. To close this, do the following:

    • Type Ctrl-c in the terminal where you did the up command.

    • Run this command to remove stopped containers:

    docker compose -f $AMR_TUTORIALS/ros2_openvino.tutorial.yml down

How it Works#

All of the commands required to run this tutorial are documented in:

01_docker_sdk_env/docker_compose/05_tutorials/ros2_openvino.tutorial.yml

To use your own image to run semantic segmentation:

  1. Copy your image into the AMR_containers folder at:

    cp <path_to_image>/my_image.jpg 01_docker_sdk_env/docker_compose/05_tutorial/param/

  2. Edit 01_docker_sdk_env/docker_compose/05_tutorials/ros2_openvino.tutorial.yml, at line 34, adding the following command:

    sudo cp ${CONTAINER_BASE_PATH}/01_docker_sdk_env/docker_compose/05_tutorials/param/my_image.jpg ../ros2_ws/src/ros2_openvino_toolkit/data/images/

  3. Edit 01_docker_sdk_env/docker_compose/05_tutorials/param/pipeline_segmentation_image.yaml to change the input_path:, line 4:

    input_path: /home/eiforamr/ros2_ws/src/ros2_openvino_toolkit/data/images/my_image.jpg

  4. Run the automated yml:

    docker compose -f 01_docker_sdk_env/docker_compose/05_tutorials/ros2_openvino.tutorial.yml up

Expected result: Execution of semantic segmentation on the image you selected

Troubleshooting#

For general robot issues, go to: Troubleshooting for Robot Tutorials.