[Paparazzi-devel] Vision Based Navigation

[Christopher Hall]

Hey All,
I would like to use a Paparazzi based UAV to achieve vision-based navigation.  For example I would like to be able follow a road or a river or a moving car based on outlines or different things.  Right now all of our image processing is done on a laptop separate from the GCS, which communicates with the GCS using the IVY bus.  Our initial idea has been to adjust a waypoints position using geo-coordinates determined from the images.  Has anyone else previously done vision-based navigation?  Is there a better way to achieve to handle in-flight re-tasking based on the camera information.  Any ideas or information would be greatly appreciated.
Thanks,
Chris.