[Tristan Lall]

Originally Posted by Andy Baker

The intent of this rule, if I understand it correctly, is to not allow metal treads, as on TechnoKats and Wildstang robots in 2002. Also, it is aimed at not allowing file cards, which were also made famous by Team Hammond in 2002, and then copied by many teams that same year. Also, there have been teams who put metal or plastic cleats or even sandpaper on wheels to give them a traction advantage. These additions were legal, back in 2002 and before, but FIRST created a rule (R08) due to too much damage being done to the carpet and field (heck, carpet would even buckle up in some places). Maybe this rule can be scrutinized to death so that no metal can ever touch the carpet, but I seriously think that is not the intent.

Andy, I agree with your interpretation of the intent. However, relying upon the eight-year-old intent of the rule writers isn't a sustainable way to codify the rules for the FIRST teams of today. (Indeed, as Gary notes, a plain reading of the rules seems to be pretty clear.)

188 also used metal treads in 2002. In Woburn's case, they were made from slices of thin-section aluminum C-channel extrusions, screwed to a turned plywood core. Having played with those wheels, I think it's fair to say that with attention to detail, they're not going to automatically ruin the carpet. Rounded corners are a must—not just for the sake of the carpet, but also so that you don't dig in when turning.

In fact, they've got the unusual feature of slipping very little on carpet in the direction of motion. This means that you can't spin your drivetrain so fast that you melt the patch of carpet below you when immobilized—compare that to conveyor belt treads which have been known to do exactly that (e.g. 188 in 2004 with some sort of wedgetop and a 6-motor drivetrain in high gear). It's not so good for your motors and your gearboxes, but it's actually better for the surface of the carpet in that circumstance. Also, because the screw heads are deeply recessed in the channels, it's harder to snag a loop of carpet and drag it with you.

However, because the carpet isn't particularly well secured, all of that traction can stretch or detach the carpet from field elements. But this isn't a feature of metal wheels—it's a feature of high-traction wheels.

Originally Posted by <R08> from 2010

ROBOT wheels, tracks, and other parts intended to provide traction on the carpet may be purchased or fabricated (“traction devices” include all parts of the ROBOT that are designed to transmit any propulsive and/or braking forces between the ROBOT and the FIELD). In no case will traction devices that damage the carpet or other playing surfaces be permitted. Traction devices shall not have surface features such as metal, sandpaper, hard plastic studs, cleats, or other attachments. Anchors (i.e. devices that are deployed/used to keep one’s ROBOT in one place and prevent if from being moved by another ROBOT) shall not use metal in contact with the carpet to “stay put.” Gaining traction by using adhesives or Velcro-like fastener material is not allowed.

I have many comments about this rule (and the similar ones from 2009 and other previous years):

1. The 2009 rule (which was less precise) and ruling was that devices that created thrust (e.g. a fan) with a component normal to the playing surface were traction devices. That is a bit arbitrary and inconsistent—any normal force (weight, change in momentum, buoyancy, etc.) contributes to the traction of the robot. The key is the traction at the wheel—nothing else even bears consideration in the traction device rule.
   
   Under the (better) 2010 rule, there is still the possibility that someone could rule that a propulsion fan—which disturbs the air impinging upon the field and exerts a measurable force upon (for example) the alliance station wall—is a "traction device" under the definition presented. (That is, unless "directly transmit" was the intended meaning—in which case, the rule should have said so, especially given the confusion about the meaning of the 2009 rule.)
   
   
2. "In no case will traction devices that damage the carpet or other playing surfaces be permitted." This implies a mandatory consequence of damaging the field, but not necessarily a prohibition of devices that are likely to damage the field (but haven't yet done so). There are other rules (like <G24> and <G26>) that specify in-match penalties for field damage, but again, they happen after the fact.
   
   This isn't a bad thing—it makes sense to know whether an item is going to damage the field, before banning it and inconveniencing the team. But when enforcing that provision in this way, it would be helpful to know that FIRST has the same understanding of the rule—and that they'll put up with one match or practice match worth of field damage, in order to prove a violation.
   
   
3. The sharp edge rule (<R05>) should also include potential field/venue damage—that would pre-emptively deal with most obviously deficient designs, without actually risking damage.
   
   
4. The real problem with the rule is in the surface features clause. Put simply, rivets or screws holding the treads down are, by almost any definition, a surface feature. Some teams used to go to great pains to recess their screw heads into the tread, but this is rare now, given the acceptance of designs which simply use rivets or slightly-protruding screws to secure the tread. FIRST, the teams and the inspectors have all tacitly acknowledged that there's not really anything wrong with that approach, except for the fact that it violates the letter of the rule. I haven't seen this issue pressed in many years. So if that's the rough consensus, why isn't there an exception in the rules for fasteners like these? Shouldn't the same principle of no harm, no foul be in order here?
   
   
5. Similarly, "metal, sandpaper, hard plastic studs, cleats, or other attachments" are not necessarily going to cause damage. Are they therefore banned on account of providing too much traction? (Even then, the specifics of the design would seem critical to that determination.) A reason should be provided (even if it is "because we say so").
   
   
6. Isn't the anchor clause redundant? (Braking being retardation or inhibition of motion—like with an anchor.) Why even introduce the extra definition into the rules?
   
   
7. The adhesive and hook-and-loop fastener clause really belongs with the other prohibited surface features. And why (potentially) ban a dried adhesive that doesn't stick or leave residue (e.g. cast epoxy), or the loop side of the hook-and-loop fastener (pointless, but harmless)?
   
   
8. The definition of "traction device" relies upon what the device is designed to do, not what it does. That makes it much more difficult to call, in the case of mechanisms other than high-traction wheels. (Actually, it says "includes", which could mean that there are other, unspecified traction devices.) Consider the case of drive chains on bump-climbing kitbots from 2010: were those designed to aid in climbing the bump? Who do the inspectors ask—Andy & Mark, because it's their kit, or the team that put it together? And one might make the argument that an unmodified file card was designed by the engineer at the file card company—and he certainly didn't design it as a traction device.
   
   
9. The statement "may be purchased or fabricated" ought to be in a footnote or clarification box, if anywhere. (It's not really relevant for 2011, since the anomalous 2009 rover wheels are long forgotten.)

If I had my way, I'd probably phrase it something like this:

<G?> [Disablement and point penalties for field or venue damage during gameplay.]

TRACTION FEATURE – A ROBOT feature that, during GAMEPLAY, directly (or through any GAME PIECE(S)) exerts a substantial tractive or resistive force upon the PLAYING FIELD, even if not designed for that purpose.

Phrased as "feature" rather than "device", to emphasize that it's not necessarily the whole wheel that is the problem—probably just a part of it. "Substantial" is a carefully-chosen ambiguity—it could be defined with a force specification, or left to the officials' discretion; it also depends on the way that game pieces are held in that particular game. Game pieces are included here, because it's conceivable that a robot could use a game piece to enhance its traction.

PENETRATING FEATURE – A ROBOT feature that, during GAMEPLAY, penetrates the entire thickness of intact PLAYING FIELD carpet, and/or abrades or otherwise damages the substrate supporting the PLAYING FIELD, even if not designed for that purpose. Also, a ROBOT feature likely to accomplish same.

Note that traction feature defined only in terms of actual function; penetrating feature also defined in terms of likely effect, so as to make it effective at inspection (irrespective of whether the robot has been on the field).

<R?> [Sharp edges rule with PENETRATING FEATURE ban to filter out obvious problems at inspection.]

<R?> If, in the opinion of the head referee or lead robot inspector, a TRACTION FEATURE has repeatedly or severely damaged the PLAYING FIELD, any GAME PIECE or other ROBOT (either at this event or any past events), the robot must be modified to mitigate the risk of further damage, and the modifications approved by a robot inspector, before the robot is allowed to take the field.

You should avoid wheel treads that will quickly melt or wear away the field carpet when your robot spins its wheels in place. Robots with too much traction can cause the carpet to stretch or bunch up, damaging the field. Protrusions on your wheels can get caught in the loops of the field carpet, tearing it apart and entangling your robot. FIRST recommends that you test your design before arriving at the competition.

Note that this would apply to things that aren't designed for traction, but exert it anyway. Past events are included, so that if Redabot damages the field at the Magnolia regional (and it's not corrected, for whatever reason), the officials at the Championship can rely on that fact to impose the prohibition immediately.