Suggestions for resolving stuck balls

[jvriezen]

In Team Update #1, the GDC has indicated they are working on how to deal with stuck balls. Let's crowd source some ideas for how to resolve it.

My idea is that if the (head) ref deems that a robot is holding a ball and is not actively making meaningful progress in playing the game (passing for assist, truss toss, scoring attempt) after a reasonable amount of time (5 seconds?), the ref can declare that ball 'dead' and a new ball can be introduced. The original ball is now just field debris and any scoring done by it will be manually deducted. Any ball coming to rest on the truss is also declared dead immediately.

This would cover the cases of dead/disabled bot, stuck ejector as well as the hopefully unlikely 'throwing the match' scenarios.

If a ball is stuck in a bot and the ball declared dead, that bot can then still 'herd' or even carry a new ball (however unlikely) without penalty in order to gain assists.

[jee7s]

Originally Posted by jvriezen

Any ball coming to rest on the truss is also declared dead immediately.

That's potentially problematic. Say this scenario occurs and the new ball is introduced. But, at a later time the ball on the truss then experiences an outside force (another ball, a vibration, a strong breeze) and ceases to be at rest. Then the ball falls onto the floor and there are two balls of that color in play. How do you know which is the legal ball?

I'd say a nice beveled or round plastic cover for the truss that prevents the ball from being at rest on it is a safer bet. It removes the possibility of the second ball being introduced, it's cheap, and it's easy to design.

IMHO, any scenario that results in a second ball being introduced to the field without the existing ball being removed or obviously non-functional (deflated, etc) places too much of a burden on the field personnel and alliance members for tracking which ball is live.

[Nuttyman54]

I think the robot needs to be e-stopped or disabled by the field in order for the ball to be declared dead, since a radio or cRIO reset could result in the robot becoming active at a later point if not expressly disabled. This doesn't address if the ball comes loose after the disable (eg a mechanism sags, or a forceful hit on the ball later).

The issue is not just how to determine when a ball is "dead" but also deal with what happens if the ball returns to play, since the game is only designed to track one ball at a time in teleop.

I think with how rarely a ball will get stuck on top of the truss, the best recourse is to declare a field fault, similar to jammed nets in 2012.

[Bob Steele]

I think the best solution would be to allow the team with the stuck ball to E-stop their robot. Then the ball would be field debris.

I think that there should be some kind of penalty on the robot that has a ball stick... not necessarily a point penalty but a playing penalty.

If the rules are left as they are the offending alliance effectively has the "death" penalty of not being able to score any cycle balls. This is pretty harsh.

By only allowing a team with the stuck ball to e-stop, the penalty is not being able to do a triple assist and the loss of a robot during play.
This would be similar to results in previous game to the loss of a robot.

It is also similar to the results of having a robot that does not have a ball stuck but still cannot function on the field.

The Alliance can still go on and function as a two robot alliance and try to overcome that deficit.

The other issue is a ball stuck on the TRUSS.
While this would be a relatively rare situation, it does need to be covered in the rules. A ball stuck on the TRUSS is more a chance act than a robot getting stuck with a ball and should not necessarily entail any kind of penalty.

The only issue is that the ball could easily become dislodged during play AFTER it was considered field debris. This could complicate things on the field .
Considering that this should be a very rare occurrence. I would recommend a 10 second count and then some kind of signal from the head referee that would restart the team with a new cycle ball. This would be the same signal given for a standard deflated ball. If there were still other autonomous balls on the field. The pedestal would not light up.

NOTE:
I do think Evan's suggestion of declaring a field fault might be a better answer to the ball stuck on a TRUSS issue. That would be clean and would not result in the kind of confusion that that ball coming back into the field would cause. The other option would be to declare a field fault WHEN the ball falls off the truss. This situation could be strategic though because if a team were losing they could INTENTIONALLY try to knock the ball off. I think the cleanest solution is an immediate FIELD FAULT.... I don't think this would be a common event.

[themccannman]

Originally Posted by jee7s

That's potentially problematic. Say this scenario occurs and the new ball is introduced. But, at a later time the ball on the truss then experiences an outside force (another ball, a vibration, a strong breeze) and ceases to be at rest. Then the ball falls onto the floor and there are two balls of that color in play. How do you know which is the legal ball?

I'd say a nice beveled or round plastic cover for the truss that prevents the ball from being at rest on it is a safer bet. It removes the possibility of the second ball being introduced, it's cheap, and it's easy to design.

IMHO, any scenario that results in a second ball being introduced to the field without the existing ball being removed or obviously non-functional (deflated, etc) places too much of a burden on the field personnel and alliance members for tracking which ball is live.

I like this idea for the truss issue.

My suggestion for stalled/disabled/non-functional robots:

If there is a disabled robot on the field that controls their alliance's only ball that team may attempt to reconnect to their robot or dislodge the ball for a certain amount of time (e.g. 10-20 seconds). If the time expires and the robot is still non-funcitonal the refs will hit the e-stop on that robot and declare the ball as field debris at which point a new ball may be entered in to play.

If the ball later becomes dislodged from the robot I would just restart the game at that point. I doubt it would be a very common occurrence that a robot would get disabled with an alliance ball in possession, their teammates were unable to dislodge the ball, and then they later drop the ball during the match.

[billbo911]

Originally Posted by Nuttyman54

I think the robot needs to be e-stopped or disabled by the field in order for the ball to be declared dead, since a radio or cRIO reset could result in the robot becoming active at a later point if not expressly disabled. This doesn't address if the ball comes loose after the disable (eg a mechanism sags, or a forceful hit on the ball later).

The issue is not just how to determine when a ball is "dead" but also deal with what happens if the ball returns to play, since the game is only designed to track one ball at a time in teleop.

I think with how rarely a ball will get stuck on top of the truss, the best recourse is to declare a field fault, similar to jammed nets in 2012.

+1 on every point.

[Daniel_LaFleur]

I'm not so sure the rules need to be changed.

As far as a ball stuck in a robot: I believe that this is a robot design issue, and that robots can be designed to release a ball if disabled, loses power, loses comms, etc... and that those that are not designed to do so should be less desirable as a partner.

As for the Truss: This is a chance you take for the ability to get an extra 10 (or 20) points. Design you robot to ensure you clear the truss easily and practice, practice, practice (so as to not shoot when you are not in a good position).

JMHO

[AndyBare]

My suggestion would just say any ball that cannot be put back into play is "dead." No ball will replace it. This will increase downtime between cycles as an alliance lose their balls to "death" by getting them stuck, etc. It will make teams sure that they do not want to be the problem. This will encourage teams to make sure batteries are charged, autonomous works, etc, without blindly putting a robot onto a field where there is a chance of something going wrong.

[AndyBare]

To elaborate, we know each team has 3 balls that cycle in and out of the field. while 1 is in play, 1 is waiting on the pedestal, and 1 is beside it. As the field ball is scored, the next is brought in and the positions switch. A dead ball will be replaced by the next in line on the pedestal, but no ball will replace it as a third "live" ball. The alliance would then only have 2.

[Jon Stratis]

For a ball stuck in a robot...

If the robot is still functional and driving around, then I think it's really just too bad for the alliance. Yes, it sucks and you might lose the match because of it, but it was the alliance's decision to use that robot as part of the cycle. There's no way for the refs to be able to tell if a ball is permanently stuck, or if the robot will manage to eject it later in the game. Personally, I really don't want to force the refs to make a judgement call in this situation. If it does come down to a judgement call, there will be plenty of times where the team's judgement is different from the refs and people won't be happy. Can't you see the complaining now? "Why didn't you declare the ball dead 30 seconds earlier? We would have won the match!" "Why did you declare the ball dead? We just weren't quite lined up perfectly to shoot yet!"

If a robot is e-stopped with a ball, then it's easy for the ref's to call the ball dead. We don't want to encourage teams to perform full-speed ramming actions on a dead robot in order to try to eject the ball! A dead robot is probably dead for a reason, and that type of abuse won't help. I think the e-stop solution is the best one. It gives the alliance the option (in the case of a momentary loss of connection, like a cRIO reboot or a radio reset) to keep the ball in play or lose the robot for the rest of the match to get a new ball.

[Daniel_LaFleur]

Originally Posted by AndyBare

To elaborate, we know each team has 3 balls that cycle in and out of the field. while 1 is in play, 1 is waiting on the pedestal, and 1 is beside it. As the field ball is scored, the next is brought in and the positions switch. A dead ball will be replaced by the next in line on the pedestal, but no ball will replace it as a third "live" ball. The alliance would then only have 2.

Please state the rule that shows that there is ONLY 3 balls that cycle.

I do believe you are in error there.

[ToddF]

I liked the aspect of the game that required teams to come up with a technical solution to the ball-in-a-dead-robot problem. But, since they are considering changing the rules, I'll make some suggestions.

Balls stuck on trusses are a field fault.

Change autonomous mode to be 15 seconds long. For the first 5 seconds, neither goal is hot. For the next ten seconds, the goals function as before, randomly becoming hot. This allows time for floor loading of balls, and robots do not need to be preloaded. It also makes it clear which teams respect their partners enough to floor load rather than preloading their robots for auto.

During teleop, if a team feels their robot is dead, they may hit the estop. If the ball is released from the robot within 10 seconds, no other action is necessary, and their only penalty results from potential loss of alliance points. If the ball remains stuck after 10 seconds, that ball is considered dead, and the pedestal light is lit, allowing a replacement to be introduced. That team is issued a yellow card.

[ATannahill]

Originally Posted by Nuttyman54

I think with how rarely a ball will get stuck on top of the truss, the best recourse is to declare a field fault, similar to jammed nets in 2012.

I disagree.

In 2012, the net and following lexan were responsible for moving the balls out of play quickly and allowing more balls to be scored. If the net was jammed, then the field was not functioning properly.

The truss is responsible for sitting there and being a truss. If it stops sitting there or being a truss and it affects the match, a field fault should be called.

If a ball gets stuck on the truss because the truss sat there and stayed a truss, it is not a field fault.

[notmattlythgoe]

I think the E-Stop option for ball's stuck in the robot are the best option. Once the robot is E-Stopped another ball is place into play and the assist points are reset. If somehow the first ball makes its way back into play the first ball in play earns any assist points that were currently collected. Until the next ball is scored, truss and assist points are not available but the ball can score goal points. This should make it simpler on the referees in the unlikely case that a second ball comes into play later in the game.

[AndyBare]

Originally Posted by Daniel_LaFleur

Please state the rule that shows that there is ONLY 3 balls that cycle.

I do believe you are in error there.

3.1.2 MATCH Logistics
Although an ALLIANCE may start a MATCH with up to three (3) BALLS, the PEDESTAL will only be illuminated when the last BALL that started the MATCH is SCORED, effectively reducing the number of BALLS in play per ALLIANCE to one (1).