Autonomous using camera?

[greatman05]

Hello. My team wants to have an autonomous mode where the robot searches for a ball, runs to it, searches for a target, and then kicks it. I have no idea how to do this, since I have no experience in vision processing whatsoever. I read the NI tutorial on vision processing, but I really don't understand it. Also,
what components do I need to have besides the camera for something like this to work?

[Greg McKaskle]

The default project contained some of this code, the stuff to identify and rotate to face the target. There have been some discussions about finding the balls using the camera, but the default code doesn't implement it. In fact, it is likely that the camera mount will not be able to see the ball and target at the same time. You may want to look for other ways of locating a ball as well.

Greg McKaskle

[greatman05]

Well then, what are those methods? I'm completely unsure as to how to approach this.

[Alan Anderson]

You can place your robot and your balls in known locations. The autonomous program doesn't need to seek out a ball if it can assume a ball is in a specific spot. All you need to do is drive the robot to that spot consistently.