Install either the deb or apt rpi-imager release.

1. deb
   Download deb file
   

   $ cd Downloads
   $ sudo dpkg -i imager_[you_rversion]_amd64.deb #check the file name downloaded
   

   If you have any dependency errors, use the following command to force the install.

   $ sudo apt-get install -f
   $ rpi-imager
   

2. apt

$ sudo apt install rpi-imager
$ rpi-imager

1. Edit the SSH configuration files
   [TurtleBot3 SBC]

   $ sudo nano /etc/ssh/sshd_config.d/50-cloud-init.conf
   

   

2. Install net-tools and check your ip.
   [TurtleBot3 SBC]

   $ reboot
   $ sudo apt update
   $ sudo apt install net-tools
   $ ifconfig
   

   

3. Enter the command below on the remote PC and use your password for the Ubuntu system.
   [Remote PC]

   $ ssh ubuntu@{IP Address of Raspberry PI}
   

• Create 2GB swap memory. [Remote PC]

  $ sudo fallocate -l 2G /swapfile
  $ sudo chmod 600 /swapfile
  $ sudo mkswap /swapfile
  $ sudo swapon /swapfile
  

• The following command ensures that the swap file is automatically activated when the system is rebooted.

  $ echo '/swapfile none swap sw 0 0' | sudo tee -a /etc/fstab
  

• Check swap memory.

  $ free -h
  

  

1. Install camera-ros, ros-humble-image-transport-plugins, v4l-utils.
   [TurtleBot3 SBC]

   $ sudo apt-get install ros-humble-camera-ros ros-humble-image-transport-plugins v4l-utils
   

   • camera-ros: A package that publishes camera output as a topic.
   • ros-humble-image-transport-plugins: Converts image_raw to compressed images for smoother transmission.
   • v4l-utils: A utility that assists with connection.
2. Run camera_node.
   [TurtleBot3 SBC]

   $ ros2 run camera_ros camera_node --ros-args -p format:='RGB888' -p width:=640 -p height:=480
   

3. The error message Unable to open camera calibration file [/home/ubuntu/.ros/camera_info/imx219__base_soc_i2c0mux_i2c_1_imx219_10_640x480.yaml] will appear if the camera calibration file is missing. After performing camera calibration, place the corresponding info file in the specified folder.
   The camera_name should be set as imx219__base_soc_i2c0mux_i2c_1_imx219_10_640x480
   Calibration yaml file example

   image_width: 640
   image_height: 480
   camera_name: imx219__base_soc_i2c0mux_i2c_1_imx219_10_640x480
   frame_id: camera
   camera_matrix:
     rows: 3
     cols: 3
     data: [322.0704122808738, 0, 199.2680620421962, 0, 320.8673986158544, 155.2533082600705, 0, 0, 1]
   distortion_model: plumb_bob
   distortion_coefficients:
     rows: 1
     cols: 5
     data: [0.1639958233797625, -0.271840030972792, 0.001055841660100477, -0.00166555973740089, 0]
   rectification_matrix:
     rows: 3
     cols: 3
     data: [1, 0, 0, 0, 1, 0, 0, 0, 1]
   projection_matrix:
     rows: 3
     cols: 4
     data: [329.2483825683594, 0, 198.4101510452074, 0, 0, 329.1044006347656, 155.5057121208347, 0, 0, 0, 1, 0]
   

1. Install ros-humble-v4l2-camera, raspi-config, ros-humble-image-transport-plugins, v4l-utils.
   [TurtleBot3 SBC]

   $ sudo apt-get install ros-humble-v4l2-camera raspi-config ros-humble-image-transport-plugins v4l-utils
   

   • ros-humble-v4l2-camera: A package that publishes camera output as a topic.
   • raspi-config: A tool for configuring camera device connections on Raspberry Pi.
   • ros-humble-image-transport-plugins: Converts image_raw to compressed images for smoother transmission.
   • v4l-utils: A utility that assists with connection.
2. Run raspi-config. v4l2-camera package uses the legacy driver. So we must configure the use of the legacy driver.
   If this step is completed, the camera node of the camera-ros package will no longer be able to detect the camera. To use the camera-ros package after this step, you must disable the legacy driver again.
   [TurtleBot3 SBC]

   $ sudo raspi-config
   

   Select Interface Options.
   
   Select I1 and set enable legacy camera support. This allows the use of the legacy driver, bcm2835 MMAL.

3. Create a YAML file containing calibration information. You can create the file using the camera_calibration package. (Currently, this step is skipped in the manual.)
   [TurtleBot3 SBC]

   $ nano ~/calibration.yaml
   

   image_width: 640
   image_height: 480
   camera_name: your_camera_name
   frame_id: camera
   camera_matrix:
     rows: 3
     cols: 3
     data: [322.0704122808738, 0, 199.2680620421962, 0, 320.8673986158544, 155.2533082600705, 0, 0, 1]
   distortion_model: plumb_bob
   distortion_coefficients:
     rows: 1
     cols: 5
     data: [0.1639958233797625, -0.271840030972792, 0.001055841660100477, -0.00166555973740089, 0]
   rectification_matrix:
     rows: 3
     cols: 3
     data: [1, 0, 0, 0, 1, 0, 0, 0, 1]
   projection_matrix:
     rows: 3
     cols: 4
     data: [329.2483825683594, 0, 198.4101510452074, 0, 0, 329.1044006347656, 155.5057121208347, 0, 0, 0, 1, 0]
   

4. You can check camera_name by this command.
   [TurtleBot3 SBC]

   $ v4l2-ctl --list-devices
   

   In this case, camera name is mmal_service_16.1.
   

5. Run v4l2_camera_node.
   [TurtleBot3 SBC]

   $ ros2 run v4l2_camera v4l2_camera_node --ros-args -p image_size:=[640,480] -p camera_info_url:="file:///home/ubuntu/calibration.yaml" -p output_encoding:="yuv422_yuy2"
   

   output_encoding parameter refers to the image encoding type. The default value is rgb8, but since it involves a conversion process, it can be slow.
   Therefore, it is recommended to set it to “yuv422_yuy2” and perform the conversion on the remote PC after subscribing to the topic. In this case, the image_compressed topic cannot be viewed directly using rqt. So if you want to view the camera using rqt, change it to rgb8.