Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Rules/Strategy (http://www.chiefdelphi.com/forums/forumdisplay.php?f=6)
-   -   Defensive wedge done differently (http://www.chiefdelphi.com/forums/showthread.php?t=118424)

Andrew Schreiber 14-08-2013 17:32

Re: Defensive wedge done differently
 
Quote:

Originally Posted by Lil' Lavery (Post 1287251)
A sensible and common goal. But I'd suggest watching some more match footage from as many different games as you can, as this goal doesn't really align with reality all that often.

While a mechanical brake will help you maintain position, in most cases there's enough room for a skilled offensive robot to avoid a fixed obstacle (like a team braked in position). Occassionally there are chokepoints (like next to the pyramids this year or the tunnel in 2010), but there are usually other options (under the pyramid this year or over the bumps in 2010) and most of the best offensive robots are capable of using those. If anyting, a braking device will be most helpful for maintaining offensive positioning during a lengthy scoring process in an unprotected area (such as most teams scoring tetras in 2005, tubes in 2007, or a bunch of balls in a single load in 2006).

Similarly, the concept a omni-directional offensive robot seems like it would more difficult to defend, but reality tends to disagree. Unless you have the option of goals in multiple directions (2005), the general path the offensive robot is going to take is predictable, regardless of how many different directions the robot can travel. A smart defender is usually able to position themselves between the offensive robot and their destination and force the offensive machine into at least some contact (a scenario that does not benefit mecanum or omni wheels). If you want to avoid this contact, the solution is typically having a higher acceleration and better drivers than the defender.

Building off of this, a smart defender knows full well that they don't need to stop their victim.They must merely slow them down enough that their scoring output is lowered past the potential scoring output of the defender during that time. Otherwise the defense is not effective.

The alternate option is that your job as defender is to disrupt the pattern of the opposing drivers. However, a skilled driver can factor a static obstacle into their pattern and work around it. Example: Most cyclers this year had a path they liked to take from feed to fire. Disrupting this path would slow them down as they then had to reorient themselves and find other reference points to line up for. A static obstacle (braked robot) is pretty useless.

Additionally, I would assert that the moment the opposing robot hits you you've done a good deal as the defender. Why? momentum is important. If I can force a robot to have to accelerate up to speed again I am delaying them far more than the contact time. Plotting your offensive routes such that you cross their typical paths before they do and force their drivers to slow down to avoid you means you can play offense while still denying paths to your opponents.

TL;DR - Less man to man defense more zone defense.

(I seriously hope any of that made sense)

Kevin Leonard 14-08-2013 18:05

Re: Defensive wedge done differently
 
A mechanical brake could have been applied this year to help defend full-court shooters. Get into place in front of them, then apply the mechanical brake to prevent an enforcer from pushing you out of the way.
This kind of makes me think we had thought of that for IRI when we tried to defend HOT in our first qual match. We could have blocked them, and not broken our drivetrain. It would have been so much better.
But looking at this game at kickoff day, I never would have thought a mechanical brake could be useful at all.
In 2012 fender shooters could have used them to stay in place.
In 2010, blocking the tunnel and/or a goal.
In 2007 for scoring on the rack.

But clearly it's not made for every game, or even for every role in every game.
I think I'll keep this "defensive wedge-brake-thing" in mind when kickoff day comes. :D

Dominick Ferone 14-08-2013 18:29

Re: Defensive wedge done differently
 
Quote:

Recently my team has been discussing ideas of what kind of drivetrain we might want to use next season. One discussion points that has often come up is traction. Wheels such as omni and mechanum provide great maneuverability but lack traction and usually void you of most defensive capabilites despite the setup of the rest of your drivetrain. Assuming you havn't built a wheel switching mechanic (that I find would likely take up a lot of space, weight, and time), it's difficult to be both very pushy and agile
My team this year had a drive train with omni wheels in the back and plaction wheels in the front so we can turn easily but also be able to have some traction. Our programmers using our wheel encoders made it while we hold down a button we can't be moved forward and back and did it for the left and right swing with the gyro.
We would go to the unprotected feeder and people wouldn't be able to push us sideways we actual broke a teams drive train while the tried to push us.

Team3266Spencer 14-08-2013 21:21

Re: Defensive wedge done differently
 
Quote:

Originally Posted by Lil' Lavery (Post 1287251)
A sensible and common goal. But I'd suggest watching some more match footage from as many different games as you can, as this goal doesn't really align with reality all that often.

While a mechanical brake will help you maintain position, in most cases there's enough room for a skilled offensive robot to avoid a fixed obstacle (like a team braked in position). Occassionally there are chokepoints (like next to the pyramids this year or the tunnel in 2010), but there are usually other options (under the pyramid this year or over the bumps in 2010) and most of the best offensive robots are capable of using those. If anyting, a braking device will be most helpful for maintaining offensive positioning during a lengthy scoring process in an unprotected area (such as most teams scoring tetras in 2005, tubes in 2007, or a bunch of balls in a single load in 2006).

Similarly, the concept a omni-directional offensive robot seems like it would more difficult to defend, but reality tends to disagree. Unless you have the option of goals in multiple directions (2005), the general path the offensive robot is going to take is predictable, regardless of how many different directions the robot can travel. A smart defender is usually able to position themselves between the offensive robot and their destination and force the offensive machine into at least some contact (a scenario that does not benefit mecanum or omni wheels). If you want to avoid this contact, the solution is typically having a higher acceleration and better drivers than the defender.

Well in that case it will be a great offensive benefit. Nevertheless I think the idea has potential and I'd like to prototype some variations before next season and maybe one will be a viable option when we know what the game is. Thank you for the advice.

Team3266Spencer 14-08-2013 21:23

Re: Defensive wedge done differently
 
Quote:

Originally Posted by Dominick Ferone (Post 1287257)
My team this year had a drive train with omni wheels in the back and plaction wheels in the front so we can turn easily but also be able to have some traction. Our programmers using our wheel encoders made it while we hold down a button we can't be moved forward and back and did it for the left and right swing with the gyro.
We would go to the unprotected feeder and people wouldn't be able to push us sideways we actual broke a teams drive train while the tried to push us.

You can still be overpowered though.

EricH 15-08-2013 01:17

Re: Defensive wedge done differently
 
I can think of twice in the last 15 years where 330 has used any form of immobilization device--that's it. (1999 had the potential to immobilize itself, but as that's more closely related to an octocanum than to a "planter", that doesn't count.)

In 2002, the game required teams to hold up to 3 goals in one zone of the field. 330 dropped the bottom of their frame--which happened to be covered in traction material--onto the carpet, tipping the robot up in the process. While we could come down, we usually didn't. Other teams did similar tricks, as I recall.

In 2003, part of the game was to hold position on top of a ramp/platform. Two wedges that locked over the edges of the platform did the trick--but they raised at the end of the match if they hadn't already done so. Again, we weren't the only ones to do something of this nature.

You'll notice that in both cases, the game essentially required the robot to be stationary for a significant period of time. The last time a game had that--besides the balanced bridges in 2012--was 2003, and most of the time that was "the first one up locks down, everybody else fights to get to the rest of the spots". There hasn't been a game since where robots have needed to be stationary outside of the end of the match for more than a few seconds at a time.

Translation: Interesting idea, but probably a waste of time for actual use unless the game requires--or your team thinks the game requires--being stationary for longer than a few seconds, or if parking on a sloped surface at the end of the match may be required. IMO, you're better off with octocanum, which is rather complex, but is only one linked system instead of two independent systems, which can make life easier.

You could also do something much simpler. Ever try to move a mecanum robot when all 4 wheels are prevented from rotating? Say, by an on-wheel brake system or a pin in a convenient hole in a sprocket?


All times are GMT -5. The time now is 02:39.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi