How to run ifm3d-ros2 from inside a Docker container
Follow these steps to get our supplied Docker container to run on your system.
Note
The instructions below apply to ROS2 Humble. Please change the commands to suit your ROS 2 distribution.
Build the docker container
We provide a Dockerfile and a build script to help you build a docker container with ifm3d and ifm3d-ros2. To get started, check out the build_container.sh script. Open it up and adjust the arguments to suit your target architecture (arm64v8 or amd64) and the targeted ifm3d and ifm3d-ros2 version. Then you can build the container:
$ ./build_container.sh
[+] Building 675.3s (23/23) FINISHED
=> [internal] load build definition from Dockerfile 0.0s
=> => transferring dockerfile: 2.92kB 0.0s
=> [internal] load .dockerignore 0.0s
=> => transferring context: 2B 0.0s
=> [internal] load metadata for docker.io/arm64v8/ros:humble-ros-core 1.1s
=> [internal] load metadata for docker.io/arm64v8/ros:humble 0.9s
=> [auth] arm64v8/ros:pull token for registry-1.docker.io 0.0s
=> CACHED [build 1/9] FROM docker.io/arm64v8/ros:humble@sha256:13eed2b61402a7be4dcfb1463398966f27fe807e81e447456d167627ce9ee8ee 0.0s
=> CACHED https://github.com/ifm/ifm3d/releases/download/v1.2.6/ifm3d-ubuntu-22.04-arm64-debs_1.2.6.tar 0.6s
=> CACHED [stage-1 1/7] FROM docker.io/arm64v8/ros:humble-ros-core@sha256:b58d5d27371fac49e2d50649cb37effa390a3f854a473689a333c44e64a66f81 0.0s
=> [build 2/9] RUN id ifm 2>/dev/null || useradd --uid 30000 --create-home -s /bin/bash -U ifm 0.5s
=> [build 3/9] WORKDIR /home/ifm 0.0s
=> [build 4/9] RUN apt-get update && apt-get install -y git jq libxmlrpc-c++8-dev libproj-dev build-essential coreutils cmake wget 172.1s
=> [build 5/9] RUN mkdir /home/ifm/ifm3d 0.5s
=> [build 6/9] ADD https://github.com/ifm/ifm3d/releases/download/v1.2.6/ifm3d-ubuntu-22.04-arm64-debs_1.2.6.tar /home/ifm/ifm3d 0.0s
=> [build 7/9] RUN cd /home/ifm/ifm3d && tar -xf ifm3d-ubuntu-22.04-arm64-debs_1.2.6.tar && dpkg -i *.deb 2.1s
=> [build 8/9] RUN mkdir -p /home/ifm/colcon_ws/src && cd /home/ifm/colcon_ws/src && git clone https://github.com/ifm/ifm3d-ros2.git -b lm_humble_tests --sin 4.2s
=> [build 9/9] RUN cd /home/ifm/colcon_ws && source /opt/ros/humble/setup.bash && colcon build --cmake- 242.8s
=> [stage-1 2/7] COPY --from=build /home/ifm/colcon_ws /home/ifm/colcon_ws 0.1s
=> [stage-1 3/7] COPY --from=build /home/ifm/ifm3d/*.deb /home/ifm/ifm3d/ 0.0s
=> [stage-1 4/7] WORKDIR /home/ifm 0.0s
=> [stage-1 5/7] RUN apt-get update && apt-get install -y --no-install-recommends libxmlrpc-c++8v5 locales sudo libssl-dev libgoogle-glog0v 238.2s
=> [stage-1 6/7] RUN cd /home/ifm/ifm3d && dpkg -i *.deb 2.1s
=> [stage-1 7/7] RUN echo "en_US.UTF-8 UTF-8" >> /etc/locale.gen && locale-gen en_US.UTF-8 && /usr/sbin/update-locale LANG=en_US.UTF-8 7.1s
=> exporting to image 3.3s
=> => exporting layers 3.3s
=> => writing image sha256:c1d72f3316c084ee0bb759c9aa9bdfdd64bcdcff33a3011efeadcf6f38815160 0.0s
=> => naming to docker.io/library/ifm3d-ros:humble-arm64_v8
Deploy the container
To deploy the container on to the VPU, or use the container locally to interact with the O3R platform, please refer to our docker documentation.
Resource management
Resources on the OVP8xx are limited and shared between all the running processes. We recommend assigning the Docker process to specific cores so as not to interfere with other applications. Refer to the resource management documentation on ifm3d.com.
Distributed setup
It is possible to run a complete ROS system in a distributed way. In this section we provide instructions to run ifm3d-ros2 in a container deployed on the VPU (primary container), and the visualization locally on a laptop (secondary system). These instructions can be adapted to suit other architectural designs.
Primary docker container: ifm3d-ros2 node
Build and deploy the docker container image to the VPU (see the documentation on how to deploy a container):
$ docker save docker.io/library/ifm3d-ros:humble-arm64_v8 | ssh -C oem@192.168.0.69 docker load
SSH to the VPU and run the container:
$ ssh oem@192.168.0.69 #Adapt to the IP address of your VPU
o3r-vpu-c0$ docker run -ti --net=host ifm3d-ros:humble-arm64_v8
root@952330b98eac:/home/ifm#
Source ROS2 and the
ìfm3d_ros2installation:
root@952330b98eac:/home/ifm# source /opt/ros/humble/setup.bash && source colcon_ws/install/setup.bash
All ROS nodes should see each other as long as they are on the same ROS domain ID. The default
ROS_DOMAIN_IDis 0 and doesn’t get changed here.Run the ros2 node:
$ ros2 launch ifm3d_ros2 camera.launch.py
Note
The command above uses the default configuration defined in the /config/camera_default_parameters.yaml file. You can adapt the parameters by using a different config file:
root@62b0c2e120bb:/home/ifm/$ ros2 launch ifm3d_ros2 camera.launch.py parameter_file_directory:=config/examples parameter_file_name:=o3r_3d.yaml camera_name:=camera_3d
Second ROS2 client: plain ros2
Source ROS2 on the secondary machine:
$ source /opt/ros/humble/setup.bash
Check that ROS topics are available (on
ROS_DOMAIN_ID=0):
$ ros2 topic list
/ifm3d/camera/amplitude
/ifm3d/camera/cloud
/ifm3d/camera/confidence
/ifm3d/camera/distance
/ifm3d/camera/extrinsics
/ifm3d/camera/raw_amplitude
/ifm3d/camera/transition_event
/parameter_events
/rosout
/tf_static
Start the visualization and subscribe to the relevant topic:
$ rviz2
Tip
To use Rviz2 from a docker container, the container has to be started with a specific environment:
$ docker run -it -v /tmp/.X11-unix:/tmp/.X11-unix --env=DISPLAY --env=QT_X11_NO_MITSHM=1 --net=host --privileged ros:humble-ros-core