Gate Vision Module data

[tylerstubenvoll]

From the Github issue tracker:

Our sub must pass through a gate to make it to the semi-finals. We will have a small vision module that will aid the AI in passing through the middle as intended.

The gate specs aren't given to us in the competition rules (see yourself: http://www.auvsifoundation.org/competitions/robosub/robosubteamcentral)

But it's an incredibly simple task. I am just opening this issue for comment: Exactly what information would the AI team want, and do they have any comments on format? I intend to completely close this issue within a week, so choices will be final unless mission critical. 

Recall for this problem: There are three forward facing cameras. One directly forward, and two angled outward for field of view. We do not have exact dimensions of the gate (relative distance will be a problem / not possible) The gate was (from past descriptions) just a big square rectangle that leaned a little in the pool but extended from the bottom of the pool to the top. 

Any other comments?