To connect to cRIO: place pandaboard running linux on robot to interface between the kinect and the cRIO. Program the pandaboard to do all of your image processing, outputing only a x,y and distance value to the cRIO. from other threads it seems like you should get ~30 fps processed this way, depending on how intensive your code is. of course, all power connections have to be done custom.
As an extra option, do not connect the pandaboard to the cRIO, but instead get it approved to communicate with your driver station laptop via wifi, per [R67]
Quote:
[R67]
Any decorations that involve broadcasting a signal to/from the Robot, such as remote cameras, must be approved by FIRST (via e-mail to frcteams@usfirst.org) prior to the event and tested for communications interference at the venue. Such devices, if reviewed and approved, are excluded from Rule [R55].
|
If you did this, you could probably just do all the image processing on your driver station laptop, and view the image besides.