Docker (Windows 10)

ROS2-PathPlanning-Docker Setup (Windows-10):

Author: Haider Abbasi

Pre-Requisites:

• Get Docker desktop installed and running by following this Guide
• Ensure WSL-2 is installed or upgraded from WSL-1 (if you have older version).
• Install and run VcXsrv Windows X Server from this link

  Required for Graphic Applications to run on Windows (e.g. Gazebo)

  • (After Installing): Run xlaunch inside (C:\Program Files\VcXsrv)
  • Keeping Multiple Windows, Set Display number to 0, Click next
  • Click next again (Keeping default values)
  • In Extra settings: Uncheck native opengl + Check Disable access control
  • Click next and Save Configuration for running without needing to configure again.
  • Click Finish and XServer is launched (y)

Project Setup:

1. Run the docker image using the following command.

( For Cuda-enabled Graphic cards ) Faster

docker run -it --name r2_pathplanning_container -e DISPLAY=host.docker.internal:0.0 -e LIBGL_ALWAYS_INDIRECT=0 --runtime=nvidia haiderabbasi333/ros2-pathplanning-course:1 bash

( For Non-Cuda (Amd or CPU) ) Slower!

docker run -it --name r2_pathplanning_container -e DISPLAY=host.docker.internal:0.0 -e LIBGL_ALWAYS_INDIRECT=0 haiderabbasi333/ros2-pathplanning-course:1 bash

• To enter an already running container run,

docker exec -it r2_pathplanning_container bash

2. Execute Gazebo launch file

ros2 launch maze_bot maze_1_robot_camera.launch.py

3. Open a new terminal (cmd_prmpt,Windows Terminal, Powershell) then enter into the running container by executing the following line

docker exec -it r2_pathplanning_container bash

To build the project (again). Perform,

cd root/ROS2-Path-Planning-and-Maze-Solving/path_planning_ws/

colcon build

4. Execute the maze-solver node. 👍

ros2 run maze_bot maze_solver

---

Docker Commands

Useful docker commands.

1) CONTAINERS

• Stop containers

sudo docker kill $(sudo docker ps -a)

• Delete all containers

docker rm $(docker ps -a -q)

• Start a container

docker start (container_id)

2) IMAGES

• Delete all images

docker rmi $(docker images  -q)

• Removing Dangling images

docker rmi $(docker images --filter "dangling=true" -q --no-trunc)

• Connect shell to running container

docker exec -it (container_id) bash

• Remove Dangling everything (BEWARE!)

docker container prune
docker image prune

• Run container from image-ID and name it using --name.

docker run --name ros2_foxy -it cd2349caf569