We are using Dagu's tracked chassis. The body of the vehicle is 3D printed. The BeagleBoard-xM is used as on-board computer to control wide range of sensors (such as ultrasonic range finders, compass, cameras, etc.).
Regarding video - exactly, we are capturing video and then using OpenCV to detect the circle. After that, the distance between center of the circle and center of the frame is used as an error to drive PID-based control algorithm. You can check https://github.com/andreynech/udacity-cs373/tree/master/unit5 for more details.
Thanks Dan M for the positive feedback! :-)
The two cameras are used for stereo-vision, so they are looking forward. However it does not work as well as we would like it to be yet (problems with USB drivers). That is why, there is also the version with one camera.
Looks cool. Video processing to find the target... It says 2 video cameras. Are they used for triangulation, depth 3D perception?
Or is it one front, one rear for separate pictures?
Oh and I like the 3-D printed body parts, too.
Sorry, used the new comment button instead of reply :-) . Please see my message above.