[FTC]: Warning about Exclusively-Defensive Robots(and how to build one)

[Chilli]

I believe Defensive Robots will shut down nearly every top team I've seen, including Oly Cow, Masquerade, and i^2 robotics. Although you guys might have seen robots playing defense on you, the introduction of the super heavy exclusively defense robot will be a whole different story.

So, at West superregionals at least, and perhaps in other places, there will be at least one defensive robot. I feel that, if you score balls, and especially if you are a high scoring team, it's likely that you will be shut down by a team like the one that shut us down, the Enterprisers.

Just for some reference, our robot scored in a match 485 points by ourselves, as well as 535 with a partner. In our field test, allied with another robot, we scored 737 points. This was in addition to us not dragging the 2 goals back to the ramp as well as us deciding to practice on the opposing team's center goal instead of going back up the ramp. We ended up as 2nd alliance captain, winning our first series 580-~100 and 400-~200. However, as said earlier, a defensive robot, Enterprisers, shut us down effectively.

Specs: 40 pound robot, with 8 andymark stealth wheels, 4-6 motors on their drive train(don't remember exactly). However, it's likely that they'll add up to 8 motors on their drive train for supers/worlds. They do have a cup where they can place one large ball, but we never saw them succeed in scoring. Generally, they score 0 points in the match, other than coming off/on the ramp.

This is their strategy. Often, during autonomous, they'll drive into our goal, or alternately, pin us. This prevents us from scoring in the rolling goal, and puts them in perfect position for their teleop strategy.

During teleop, they use several defensive strategies.

1: They have a button that pins for 4.8 seconds and then backs out. This allows them to avoid penalties for pinning. After backing out for a bit, they ram us again, making it very difficult to maneuver. It is extremely difficult to push against them, as they are 40 pounds, have immense traction with 8 wheels, and have lots of power with 4-6 motors. This does not invalidate any rules and was not called for any penalties.

2: They block us from getting to the rolling goals. If they sit in the space between the 60, the ramp, and the center structure, they have the perfect position to block us. It isn't considered "blocking access" to the center goals as if they couldn't move, you could move around them. However, in practice, this is very difficult unless they mess up. If you shift up, they shift up, and if you shift down, they shift down. With the walls and the center structure in the way, this would be nearly impossible to get back if you executed completely perfectly.

3: If you manage to grab onto the goal, they start pushing you around. They did only follow the 90 cm robot around, but 90 cm gives a lot more points than 60 cm. Perhaps our robot, with only 2 stealth wheels and weighing only 22 pounds was especially susceptible to us, but they pushed us around like they were playing battlebots. This allows them to pin us against the wall, where they can resume strat number 1. Alternately, they've also pushed us onto the kickstand where we were stuck. Even if they can't do any of those, it makes it very difficult to pick up balls. As stated before, it's very hard to resist being pushed against this team. 8 stealth wheels gives great traction, 4-6 motors gives lots of power, and 40 pounds makes it hard for them to be pushed.

4: If they mess up and you manage to grab 5 large balls and go for a dump, they back up and ram you as you're dumping balls. Our lift was not stable enough at 90 cm to dump consistently, but I've seen their strategy also work against continuous robots.

During endgame: During endgame, they have one primary goal. Prevent teams from scoring in the 120 cm goal. To do this, they tend to skirt in the area around the center goal, nudging the other team here and there. If they can, they try to get in between you and the center goal. Otherwise, if you manage to get in position, they back up and ram you. For us, this would often push us out of position, and made it difficult to line up properly. In addition, due to the force their 40 pound robot can apply, our lift would wobble a lot.

I will note that they tipped over a robot going for 120 cm in the finals. Although they got a major penalty for this, 50 points < 162 points. Also, that other robot couldn't do anything for the rest of the match. It ended up being a net positive for them.


ABOUT PENALTIES:
In the matches that we saw, they did not get penalties for any of their actions(except for tipping over a team). They have learned their rules extremely well, and through a combination of good driving and programming(ie: their 4.8 second pin preset)

Any team that hopes to go far should think long and hard about how they would deal with a robot like this. Here are some questions you should try and answer. If you can't think of a solution, defensive robots may be a problem.

1. How will you break past the pinning cycle? As you can see in the rules, this is not prohibited at all. G10: A Robot cannot Pin or Trap another Robot for more than five seconds. As you will be able to move in the 1-2 seconds they back up, this strategy was considered legal.

2. How will you get to the 90 cm goal? You likely will not be able to push them out of the way, and it's very difficult to get near it if they play well. One thing we tried was grabbing the 60 cm and letting our lower scoring partner grab the 90. Although the enterprisers focused on the 90 cm robot, leaving us free to fill the 60, we were not able to get nearly as many points. Still, we lost, something like 340-354(due to a major penalty they forced us into committing).

3. Can you dump in the 90 while getting rammed? I'm guessing that many teams will have trouble doing this. Most lifts are not extremely stable, and the enterprisers crashing into you causes a lot of force.

4. If you can dump in the 90, can you dump in the 120 while getting rammed?

How to build a robot like Enterprisers: ~12 hours required.

I made this list so that teams can see how to build a defensive robot.

Ingredients list:
8 Neverest Motors
8 Andymark Stealth Wheels
Stuff that makes you weigh a lot.



As the Enterprisers never did use their lever for scoring, I'll forgo that part.

1. Create a super heavy chassis. I'm not quite sure how heavy you should be, but the Enterprisers were 40 pounds. Feel free to add weights to your robot to make yourselves heavier.

2. Attach your 8 motors and 8 wheels. I'm not sure what would be the best configuration of motors/gears/chains, but feel free to choose your own option.

3. ???

4. Eliminate robots that can score.

5. Make sure to read the rulebook carefully. Make sure you pin for 4-4.9 seconds, never 5.

Congratulations, you now have a very capable defensive robot. Follow the strats outlined above, and you should be ready to shut down almost every high scoring team!

To the other high scoring teams. Hopefully you guys can deal with them better than we could. Some solutions are: Adding ramps to the sides of your robot so that they're flipped upward when they ram into you, driving up the ramp so that they can't follow you.

Credit to Enterprisers for finding this strategy. I suspect it'll work against many teams.

[maths222]

A few thoughts:

1. For most teams, make your robot heavier than it is now(get some solid steel bars and bolt them on)-the heavier you are, the harder you will be to be pushed, and the easier time you will have pushing other robots out of your way, but make sure you can still turn.
2. The tipping presumable was unintentional. I will assume that it is, as otherwise that would be grounds for G19 penalty.
3. Pinning in auto: "Autonomous strategies that appear to be intended to Pin or Trap an opposing Alliance’s Robot may result in a Minor Penalty, or if chronic, Disqualification." Therefore, we will assume that their intention was simply to get in your way, and that the pinning simply happened because your autonomous changed between matches (otherwise, I don't see how that strategy would not be a G10 violation)
4. "If you shift up, they shift up, and if you shift down, they shift down." This would seem to meet the definition of trapping: Preventing an opposing Alliance Robot from accessing or escaping from a constrained Area of the Playing Field for an extended period of time." Obviously, I did not see what they were doing, but assuming that, other than eventually out-manuvering them, they moved to block any access to the rolling goal/goals, they would seem to have been trapping.
5. "They try to get in between you and the center goal" I assume that they move out after 5 seconds, as otherwise this would meet the definition of Trapping.

Overall, there are 2 main points:

• If you try to adopt such a strategy, very carefully review the rules and make sure that nothing you are doing falls into any of the seeming traps I outlined above (or some other violation). I.E. G9 (which I hope no one is violating-I have not seeing it happen, and I do not want to see it happen), G10, G19)
• If you are against a team following this type of strategy, please consider the points I raised above, and if you believe that something should have been called that was not (particularly with regards to trapping), please visit the question box. With teams hugging the border between legality and illegality, it is greatly to your advantage to read the rules as carefully as the other team.

**DISCLAIMER: I have not seen the team mentioned above play, so I may have misunderstood some aspects of their gameplay. Further, I in no way intend to suggest that they have intentionally broken any rules, or even necessarily broken any rules unintentionally.

[wgardner]

Also, from the FTC Ref Manual on page 17 under Pinning/ Trapping in Driver Controlled Mode:

"For pinning to occur, the pinned Robot must not be able to move in any possible direction. Blocking a Robot from traveling in its preferred direction is not pinning. Open Field contact between Robots is not pinning. If pinning does occur, the Referee must notify the offending Team and begin counting. The count should be visible, pointing fingers is recommended, and the count should be loud enough for the Teams to hear. Referees should start the count at 2, since by the time the count is started at least one second has gone by. The pinning Robot must back off a minimum of 36” (approx. 1.5 full Field tiles). If the offending Robot does not back up, the Referee should assign a penalty. Repeated pinning for less than 5 seconds is allowed, provided the offending Robot backs up a minimum of 36” each time. If the Robot continues to violate the Rules, the Referee should disable the Robot, but only after making the Team move the Robot away from the pinned Robot." (bold added by me)

So if they pin for 4.8 seconds, they need to back up a FULL 36", or 1.5 field tiles away. If they only back up a few inches and then start pinning again, the 5 second count should not reset at 0. And if they do this repeatedly, they should be disabled and disqualified.

I suggest that teams bring this document to tournaments in case this behavior happens.

[MattRain]

The below is just my opinion.

Over the years, we have noticed that the drive base of your robot is one of the most important things to design. Every game that has been created has had some sort of defense. To think that you will never be defended on, is just not smart, haha.

Our robots always start with a strong drive base. With the introduction to Andymark Motors, its even better now. If you look back in our last five years of the seven, every single robot has over 6 drive base wheels. Again, with the introduction to new wheels and motors, it has helped. We run the Andymark Stealth wheels since the am-2256(?) (Hi-Grip) are not allowed. Both robots also have 4 motors, 2 per side on the base. As its been said, a heavy drive base is better.... to a point.. haha. Both of the robots that we have built this year total over 45 pounds each... the heaviest robots that we have made. This is getting to the point of too much weight though. Your motors can start to struggle.

As it says in the manuals, your robot needs to be built robust for the game play, whether it be rough or not. You could hit both of our robots at full lift height, and it wouldn't be a problem for us... Even though we don't want this happening, the students have designed it in.

If you are headed to any Super Regional, (Both of my teams are headed to West), I would strong suggest looking at your drive base, and changing it to deal with the rough play.... its going to happen. You have to expect the high defense. Your in the big leagues now.

[BoRussell5037]

We actually covered defense in our latest episode of RoboTalk! (A web series we host on YouTube) If anyone wants to find out more of some of the things discussed in this thread check it out!

-----> Right Here <-----

[PHFTC]

This year's game is so full of penalties, Teams would do well to know the rules. But there are still lots of gray areas... Certainly the indicated strategy for autonomous has been ruled out in the official forums as indicated above. How do teams point out to the refs that such actions are chronic and have happened in every match the team has played in? Not sign off on the score sheet? One could be unintentional, but not 5x.

Clearly blocking access to the rolling goals is also a penalty. If we lift a tube and they hit us causing it to fall - that should be their penalty right? And we would be awarded a full tube.

I don't have an encyclopedic memory, but isn't there something about the spirit of the game/severe defense that could also come into play?

On a related note, we have seen sooo many robots with telephone cables which are way underspec (26-28AWG)- and we spent some effort working around this problem - what's the proper way to point that out to inspectors?

[MattRain]

Originally Posted by PHFTC

This year's game is so full of penalties, Teams would do well to know the rules. But there are still lots of gray areas...

Clearly blocking access to the rolling goals is also a penalty. If we lift a tube and they hit us causing it to fall - that should be their penalty right? And we would be awarded a full tube.

On a related note, we have seen sooo many robots with telephone cables which are way underspec (26-28AWG)- and we spent some effort working around this problem - what's the proper way to point that out to inspectors?

The game is full of penalties, agreed. A big defensive move that I have seen this year is when it comes to the center goal. At least in Arizona, teams have not been called for doing this. They(Teams) allow the robot to go up to the center goal and start raising, as they go to score, the oppoising alliance hits the back corner of the scoring robot, misaligning the scoring robot. The refs would allow it, as it was not blocking access to the center goal. The teams would allow the robot to get to the center goal, but then become an annoyance when it came to the scoring. (All within the 5 second rule, since its from the side, and not the back of the robot, allowing the scoring to robot to still move back and forth...)

In regards to lifting a ball tube. Its sorta an iffy area. If your running the risk of lifting a ball tube up, it should be designed with a defensive bot in mind. (I only say this, because one of my teams does lift the rolling goal enough to put onto the ramp, and they have created a lock to account for a defensive bot running into it. Most of the goal is inside of the robot. (Again its a little iffy/gray area.

On the topic of the wiring (just to derail for a minute...). At least for me, I would not call a team out through the inspectors for it. Usually a wiring issue will kill a teams chance to compete for the rest of the day, due to the fact that they cant fix the said wiring... Now saying that, it depends on where the wiring problem is. If its say the low gauge wire coming out of the battery... then that's a bit of a hazard, but when it comes to a telephone wire for a servo.. I myself would let it slide. (I have seen it. My teams all have the correct sizes... we went through the whole thing of trying to find the right wiring as well, I know how you feel.) I side track a little now, because I have called out teams in the past and this year in fact. At least for this year, there was a team that showed up, passed inspection and had about 2 hours before qualifiers started. This team was using the am-2256(?) Andymark HiGrip Wheels. When it came to these wheels, I felt that this was making the playing field un-even. I went around to some other mentors of other teams that we are very good friends with, and asked the same thing. They all agreed. We proceeded to ask the team to change the wheels out, while still notifying the inspectors, with the picture in inspection binder, that the said wheels were illegal. Something like wheels are a lot easier to replace than wiring. I say that again with looking at both sides, it really matters where the wiring problem is located.

In short, I probably would have gone over to the team, or one of the mentors for the team, and pointed out that the said wire was technically illegal, and they should change it for future competitions, higher competitions, i.e. Super Regionals and Worlds, so they wouldn't be called on it, and have to either remove or scramble to fix it.

[Sasha]

In the finals of NYC/LI championship yesterday, team 5069 (Robogamers) were quite aggressive in the endgame, repeatedly ramming our robot when we lifted our linear slide preparing to drop ball in the center goal. In the end, our robot tipped over. No penalties were given.

I can't say I wasn't angry, but I have to agree that the rules allow that.
For the superregionals, we are certainly making our robot A LOT heavier.

[JimInNJ]

Here's the bot in question. Their "defensive" conversion didn't appear to change their drive train, just beef up the pushing parts and add some weight . . .
http://youtu.be/Wzfq-caheDw

[budman]

In Texas, we had a defensive robot which simply left the parking zone, drove into all the rolling goals and then backed into their parking zone. Effectively disrupting all the rolling goals so we were unable to score any points in autonomous. Very frustrating and unGP in my opinion. We were programmed to go down the ramp, score into medium and low goals and then drag low goal to parking zone. A total of 100 points in auto. Our program took the kids a couple of practices to make reliable, theirs was probably done in 5 minutes and does not break any rules.

[MattRain]

Originally Posted by budman

In Texas, we had a defensive robot which simply left the parking zone, drove into all the rolling goals and then backed into their parking zone. Effectively disrupting all the rolling goals so we were unable to score any points in autonomous. Very frustrating and unGP in my opinion. We were programmed to go down the ramp, score into medium and low goals and then drag low goal to parking zone. A total of 100 points in auto. Our program took the kids a couple of practices to make reliable, theirs was probably done in 5 minutes and does not break any rules.

I can understand the frustration, but this is always the case every single year. One of my teams has an auto this year that dispense two balls in two goals, and brings them back, for a total of 120 points.

I personally don't see it as unGP. There is always going to be a defensive strategy in Auto or Telelop in a Match. Especially when it comes to finals, or a better alliance. Over 7 years, I have seen a lot of different strategy's in autonomous when it comes to defensive robots. My favorite robot auto, and I can't remember the team, was in the "Get over it" year. They built there auto's with a multitude of Sensors that allowed a robot to tell if something was in the way, or had gone wrong, and would then move on to a different objective. It was pretty cool to see it work. Now I'm not saying that your autonomous needs to get more advanced, but just remember that there are ways around the defensive strategy's in autonomous. There are usually signs that teams will show if they are about to do an defensive autonomous... that's your time to adjust. Make multiple different autos... that's what both of my teams do...

[ftelliott]

I sometimes wonder if the kids see the GP/uGP question the same as the adults.

As a teacher/coach and getting up there in years (when D&D was first introduced I was 40+ LOL), I have a reluctance to encourage my kids to think defensively because I value the creative process so highly and I've never played the current generation of adversarial technology games.

However, our kids grew up as "gamers" and playing against someone and thwarting their intent is completely legit as far as they're concerned. I constantly hear amazingly creative chatter about how to prevent other teams from scoring. Most of the time, I have to remind them, "That would incur a penalty." It doesn't stop the conversation, it just deflects it.

Our younger coaches play some of the kids online. Whenever one of the coaches is defeated, its a huge topic of conversation and badge of merit. Stabbing one of our coaches in the neck while online is considered cool. Hmmm...

[bduddy]

Originally Posted by budman

In Texas, we had a defensive robot which simply left the parking zone, drove into all the rolling goals and then backed into their parking zone. Effectively disrupting all the rolling goals so we were unable to score any points in autonomous. Very frustrating and unGP in my opinion. We were programmed to go down the ramp, score into medium and low goals and then drag low goal to parking zone. A total of 100 points in auto. Our program took the kids a couple of practices to make reliable, theirs was probably done in 5 minutes and does not break any rules.

Sounds effective and fair. Doing anything less would be unprofessional. Your accusations, on the other hand, are ungracious.

[nonsequitur]

Defensive autonomous is a part of the game. The disappointment, for me, is that offensive autonomous is relatively challenging for the teams to program and defensive autonomous is relatively easy. It seems too easy to defend against a team that has worked long hours to perfect autonomous. At the competitions I've attended a very small percentage of teams can detect the IR beacon and then score in the center goal. A few more - but still a small percentage - can consistently score in a rolling goal and/or bring a goal to the parking zone. Our team can drive down the ramp, score in 2 goals, and bring both goals back to the parking zone. That's 120pts and was difficult and time consuming to perfect. Contrast that with a defensive robot that takes 5 minutes to create a program to foil that 120pt effort. I guess I wish the effort required to create a defensive autonomous was close to the effort required to create an offensive autonomous.

[MattRain]

The level of autonomous building between teams differs a lot. In my first few years of FTC, yes, making an offensive autonomous would take a long time to perfect. This can be changed though. I know that if you RobotC, reach out to other teams, specially through the Robotc forums. You can build an autonomous that takes very little time to program.

I say that, because over the years, our two teams have created programming within RobotC, that allows them to create anything they want, in less than 5-25 minutes, depending on the complexity of the set program. When I sat down with my veteran team, it took them less than 3 hours to create a multitude of different programs. All of which consistently score time after time. Even my rookie team is able to create autonomous programs extremely fast, and the main programmer on the rookie team has never touch code before this year. I can walk away and wonder the pits without having to help them at all.

If you are RobotC, and would like to have a look into what the students have created over the 6-7 years, PM me. I am more than willing to show and explain it all. Its simple, modular, and still expanding to allow for more additions. (I do not know Labview, sorry)

I highly suggest looking into "advanced" Autonomous Programming and reaching out to other teams.

Defensive Strategy and Autonomous is always going to happen. Remember that. "Design for it and live with it..." that's what I tell my students.

Sorry, got a little off track...