Please login or register.

Login with username, password and session length
Advanced search  

News:

Season 13!

Author Topic: Observations from Scrimmages  (Read 4236 times)

Kims Robot

  • Let's Go Thunder!
  • Administrator
  • Hero Member
  • *****
  • Posts: 2497
  • Do or Do Not, there is no Try!
    • kimsrobot
    • Email
Observations from Scrimmages
« on: February 20, 2012, 10:51:44 AM »

There are some other observations on chief Delphi
http://www.chiefdelphi.com/forums/showthread.php?threadid=103193
Suffield Shakedown comments: http://www.chiefdelphi.com/forums/showthread.php?t=99268&page=2

Some things to think about:
1. Real bridges are harder to push down than our bridges, can our "wings" do it?
2. Hybrid and balancing appear incredibly important so far.
3. Not many (if any) teams are collecting from the protected lane, as many of the balls the human players are inbounding are rolling into the opposing alliance zone.
4. Balls are getting stuck under the ramps (on accident, no G14), FIRST may fix this, but its something to think about.
5. Getting the balls off the coopertition bridge was very helpful (some sort of hybrid that shoots all 3 balls and tips the bridge will be very powerful.
6. Balls can interfere with eachother in auto mode (most robots were shooting for 3 point shots) so some sort of delay might be useful.
7. In most matches the 2 alliance bridges were both balanced with 2 robots (essentially canceling)

I'm sure there are more... anyone else see additional observations?
Logged
~kim~
Systems Engineer - Draper Laboratory

Steve Jobs: The heaviness of being successful was replaced by the lightness of being a beginner again, less sure about everything. It freed me to enter one of the most creative periods of my life.

Josh Goodman

  • Team Member
  • Hero Member
  • ***
  • Posts: 1331
  • As time expires.
    • Facebook - Josh Goodman
    • Email
Re: Observations from Scrimmages
« Reply #1 on: February 20, 2012, 11:57:21 AM »

The successful robots so far have had an extremely quick, strong mechanism to push down the bridge. I'm not sure if the wings are on a jag or not, but it's something to keep in mind.

Consistency is key. You only have 3 shots, no time for "Oops, too high, oops too low, there we go." Applying this to our robot's "dead reckoning" or "Sans camera" mode, try to cut down as many variables as possible. So right now, position of the robot, position of the turret, and speed of the turret are all variable factors. Perhaps practicing shooting 2 points consistently from the fender might be a good "starting goal" for FLR that we can expand upon at later regionals.

Different hybrid modes are super important depending on which teams we're paired up with and how many balls are on the middle ramp. If our alliance is physically capable of storing 2 balls starting out, they should, hands down. This creates less of a bounty for teams on either side and makes it "less important" so you can concentrate on scoring in hybrid. This one crucial decision will decide a majority of qualifications matches: Tip the bridge first or shoot first? By eliminations it won't matter because the smart alliances will have bots that can do both or either or.

Defense will be flakey at best. Successful teams will find ways to annoy, defend, and obstruct teams without incurring penalties or depriving their own team. Robots that can only shoot from the fender have a huge flaw. Shorter robots though (like us) if shooting from the fender or the BACK of the key, should easily be able to outshoot any goaltending defense. Drivers have to have to have to practice field awareness and driving carefully yet swiftly around the opponents key and lane.

Field obstruction is AWFUL this year. Last year the scoring was difficult to observe but this year due to the bridges + barrier + human players + 6 robots, it's EXTREMELY EASY to lose track of any and all balls on the playing field. Coaches HAVE TO, HAVE TO, HAVE TO keep this field awareness and have an idea of where all 18 balls are at any given time.

Drivers, spend a good amount of time practicing balancing two robots on a bridge, and make sure the programmers have enough time to try and program in as many CONSISTENT auto modes as possible. If the only thing we can do is tip the bridge in auto, so be it, it's WAY important. We will not have to shoot 3 balls in the 3 point goal at FLR to be successful in auto. CT is a different story.

You will hardly see any runner bots. Due to the fact that a good trained human player can bounce a ball to traverse the field, you won't need a robot to bring balls from that corner to the scorers. You WILL see feeder bots which snag up rebounds and send them to the other side of the field. Spending time collecting balls from the other side of the field is a HUGE waste of time. If we have a successful strategy as a shooter or feeder, we shouldn't have to cross the bridge more than once during a match.

WE SHOULD NOT BE AFRAID OF DEFENSE. The pneumatic tires act as a very solid anchor device. Chances are with the limited defense and setting up a shot, we shouldn't care where we shoot from. The only flaw in our defense design where we might be able to get pushed around is in our bumper gaps. Unfortunately that does give the opposing team some vertical leverage in which to spin us. But that design is in the past and most teams won't be able to exploit it.

A lot of these piggy backed Kim's thoughts, but it really comes down to a few very simple points WE NEED TO DO to be successful.
Logged
Josh Goodman
Rolling Thunder '09
Part-Time Mentor
« e-mail redacted »