Quote:
Originally Posted by Scott Hill
Each side of the corner, side AB and side FA, is protected by BUMPERS. Each sides protection clearly meets the intent of <R08> ..."If implemented as intended, a ROBOT that is driven into a vertical wall in any normal PLAYING CONFIGURATION will always have the BUMPER be the first thing to contact the wall."... . Side AB has no BUMPERS on it yet the BUMPER configuration clearly meets the intent of the rule, which is clearly stated.
|
It looks like you're suggesting that any bumper design that satisfies the condition in the intent statement offers the required protection. That's not unreasonable, but I don't think that's the way that the GDC understands it.
As a test case for your theory, consider a robot similar to the one you posted above. Move the trailer hitch out of the way (put it in the gap on DE). Delete the section AB. Extend the segments FA and BC to their intersection, and call it G (but leave the bumpers as-is). Delete points A and B, leaving us with rigid segments GC and FG. Now, if we perform those operations on the picture above, we end up with a big sharp corner (at G) that clearly extends beyond the bumpers. This contravenes the condition in the intent statement, and is not protected by segments. But what if we made the angle at G something large, like 170°? There is still a corner (a discontinuity in the radius of curvature), but now the thickness of the existing bumpers allows it to pass the intent statement's condition (the corner doesn't stick out past the outermost edges of the bumpers anymore).
Is it your contention that even though the 170° corner at G is not abutted by any bumper segments, all necessary conditions are met (because the bumper hits first), and it would therefore be legal? If that's the case, then the amount of protrusion (inward or outward) past the edge of the bumpers is the most critical factor in determining protection (under your theory).
Apparently, the GDC considers the condition in the intent statement and the corner protection requirement to be separate, necessary conditions. Furthermore, it looks like they understand corner protection to mean a design with a legal bumper segment on each side of the corner.
Assuming that the function of the Q&A is to guide the interpretation of rules, but not to impose additional constraints*—that being the function of the rules and updates—the GDC's responses regarding corners have been mutually consistent and legal under the rules—so following them precisely ought to be acceptable at any event. (That's your best course of action.)
There might be some room for your interpretation, however: the GDC is describing a legal way to meet an existing requirement (protection of corners per <R08>, part I), but there's nothing in a rule or update that says that this is the
only possible way to protect a robot's corners—in fact, to say that there is only one legal mode of protection (without some sort of explicit definition in the rules) is a bit of a stretch of the principle that the Q&A shouldn't be defining new constraints. (Yet, I think that that's implied
here.) So as I understand it, you're relying on the fact that the next best thing to a direct definition of protection is the intent statement, and that because your proposed design passes that test, your corners are protected.
Like I said, it's not unreasonable...but you're taking a big risk that the inspectors at any given event will be open to considering that logic, and will arrive at the same conclusion as you, and that FIRST won't clarify things once and for all in an update (ruling against you).
*FIRST has not stated this directly for a couple of years, but that was formerly the rule of precedence. Maybe they made the statements in the 2009 Q&A binding, but neglected to tell anybody....