Has anyone thought about mounting both on the robot instead of just one or the other?
My team did for a short time. There are definitely some practical uses for it, but ultimately the costs are very high. For one thing, the Kinect requires an incredible amount of processing power and the cRio cannot handle it as well as a regular camera. Second, the coding to make the Kinect work at all appears to be rather involved, so ultimately we decided against it. There was another thread on this topic a little while ago…try searching for it.
the only plus can see is getting depth out of the image, which can just as easily be done with the regular camera and the ultrasonic sensor.