Revision history [back]

You can have a look at https://github.com/ros2/ros2/wiki/Design-Guide for getting a good DDS configuration for video streaming.

There is a demo: https://github.com/ros2/demos/tree/71ab3261675c1f44e1e280556d4fb323efee51f2/image_tools/src showing how to transfer images. In your case you will have a topic "video" where you have a subscriber (the server) and two publisher (the 2 Raspberry PI)

For the processing part of the images, I cannot really help you about which algorithm you should use. There are lots of website describing how to detect obstacles. I would suggest using OpenCV as a base and look for obstacle detection algorithms. https://www.intorobotics.com/how-to-detect-and-track-object-with-opencv/