E-Puck Camera

This ROS2 driver takes images from camera device located on e-puck2 robot. It uses V4L2 driver for extracting the images from the camera and Multimedia Abstraction Layer to perform fast JPEG conversion by utilizing onboard GPU.

Parameters

• quality Quality of JPEG conversion (default 8)
• interval Interval of image acquisition in milliseconds (default 80)
• width Width of the output RGB images, height is automatically calculated (default 640, valid values [640, 320, 256, 128, 64])

Topics

• /image_raw [sensor_msgs/msg/Image] Raw RGB images
• /image_raw/compressed [sensor_msgs/msg/CompressedImage] JPEG compressed images
• /camera_info [sensor_msgs/msg/CameraInfo] Information about the camera (e.g. intrinsic parameters)

There a few things to note here, mainly related to Raspberry Pi Zero computational power and WiFi bandwidth.

• Images will be published only if there is at least one subscriber to avoid unnecessary computation.
• Topic /image_raw/compressed is suitable for offboard processing since the data transmission rate is significantly lower.
• Topic /image_raw is suitable for onboard processing since no conversion is needed (much lower CPU utilization).
• If you have multiple nodes subscribed to /image_raw/compressed it would be handy to have an additional node republishing images to reduce network utilisation, e.g. ros2 run image_transport republish compressed in/compressed:=image_raw/compressed raw out:=image_raw/uncompressed.