Off-season project ideas?

[danshaffer]

We've been looking at off-season projects, assuming we can find some more money. We are already a Lemelson-MIT InvenTeam, and have a lot of work to do with that. I know a lot of teams aren't quite done with FIRST season, but what will you be doing during the off-season? I've played around with building a robot cart, maybe learning to code.
Ideas?

[Jeremiah Johnson]

Build a cart... I can't state this enough. Not saying you guys don't already have a cart, but it's huge to have a cart. Especially one that's adaptable. One that can raise and lower the robot and is extremely maneuverable.

[mizscience]

Our team hasn't officially discussed off-season projects yet as this season is not as yet finished, nut we always help run the Mission Mayhem off-season competition, so that takes up a part of the summer. We were also thinking about experimenting with mechenum [sp?] wheels or swerve drive on a practice robot, just to learn the controls and see if we possibly wanted to use them in the future.

[danshaffer]

yeah, we do have a great cart, but it doesn't raise. also i'd like to do it to just play around with pneumatics a little, although to lift the robot from say 6" to a comfortable working height like 3'6" or more would require a hell of a lot of force. we haven't used pneumatics (to the best of my knowledge) in the long long history of team 8, but to lift a bot that much would require more than most gas springs have (we should have a few shorter ones lying around, though). would a scissor lift be a good idea here?

[=Martin=Taylor=]

You guys are going to Cal-Games right?

Well, its a little-known fact that a team can enter MULTIPLE robots in Cal-Games... We talked about doing it last year, but we chose to modify our old robot instead (we added a 3pt. shooter).

Our robot doesn't need any modifying this year, so there's a good chance we'll build another robot.

We already have a duplicate drivetrain (same as the one we used this year), and a whole bunch of spare parts (including a fully functional tube manipulator). It should just be a matter of welding up a new frame and adding some ramps

We may also be using a track system at Cal Games

[Jeremiah Johnson]

Originally Posted by danshaffer

yeah, we do have a great cart, but it doesn't raise. also i'd like to do it to just play around with pneumatics a little, although to lift the robot from say 6" to a comfortable working height like 3'6" or more would require a hell of a lot of force. we haven't used pneumatics (to the best of my knowledge) in the long long history of team 8, but to lift a bot that much would require more than most gas springs have (we should have a few shorter ones lying around, though). would a scissor lift be a good idea here?

That's what we use. Works like a dream. Uses a drill to power the lift.

[danshaffer]

Originally Posted by Budda648

That's what we use. Works like a dream. Uses a drill to power the lift.

any chance we could get some pics?

[yodameister]

Originally Posted by danshaffer

We've been looking at off-season projects, assuming we can find some more money. We are already a Lemelson-MIT InvenTeam, and have a lot of work to do with that. I know a lot of teams aren't quite done with FIRST season, but what will you be doing during the off-season? I've played around with building a robot cart, maybe learning to code.
Ideas?

Contact FRC 842 about their underwater competition that will take place in June. They are located in Phoenix, so not too far from you.

[Justin M.]

Pit display, off-season events, possibly putting on our own off-season event. Mecanum prototypes for next year's chassis. Probably build an amazing cart. Combat robotics. We'll be busy up to the 2008 Kickoff most likely.

Also the freshman team is doing Sea Perch, some underwater robotics competition for middle school - freshman kids.

[CyberWolf_22]

How about an Electrical Test Box:

The test box will allow for easy debugging of electrical equipment and testing of electrical applications including motors, sensors, and batteries. An LCD will display variables in the programming including sensor values and battery voltage. Switches and pots will allow easy control and testing of devices. It will also be easily reprogrammable to allow for other applications such as autonomous motor and transmission tests. A Vex controller or an EDU bot controller would work well for this application.

[Bongle]

As someone not in high school anymore, I really wish I had a RC and OI to do this stuff myself, they're all pretty fun programming challenges.

Idea #1
If you didn't have an autonomous mode this year, make a successful one. Something I noticed is that 2006 and 2007 had very similar autonomous modes (follow the light->do game action). Having an already-done autonomous for 2006 would have vastly helped any team this year.

Idea #2
Make a general-purpose dead reckoning system with the gyro and accelerometer. Drop a RC, OI, wheel counters, and the gyro/accelerometer chips in your programmers lap and tell him to do it (that's all he needs. He can tape the chips to the RC and just move it around to test). Make it so that you can just call a single function in code like "driveTo(x,y)", and your robot will go there. If you make it so that driveTo(x,y) doesn't return (this can be done with a while loop inside the function that does your getdata/putdata) until your robot is there, then it makes for a VERY easy autonomous mode:

if(wereHittingTheOtherSideOfTheRack)
{
  driveTo(otherSideOfRackX,otherSideOfRackY) // goes to the other side of the rack
}
doAutonomous() // finds the light and score from current position

Idea #3 (to be used in conjuction with idea #2)
Make an on-robot interface that allows you to specifiy a position on-field for your robot to go to in autonomous without having to recompile the code. At first, this can just be potentiometers for x and y coordinates, but a team at nationals last year had a touchscreen display of the field that allowed them to plan an entire defensive path while they were setting up their robot on-field.

Idea #4
Insipired by pattern-running torpedoes, make a defensive autonomous mode where if it doesn't hit the target the first time, it turns around and tries again with a slightly different aim point. You can use the accelerometer and watch for spikes to detect a hit. Another torpedo-inspired autonomous mode would run a certain pattern once it got to its target point: widening circles, a mow-the-lawn pattern, etc.
Code could look like:

driveTo(rampPositionX,rampPositionY)
if(weHitObject())
{  nudgeEnemy();
}
else
{  driveTo(startPositionX,startPositionY); // makes another pass
}

Idea #5
Haha, I've been playing too many submarine games lately. Inspired by acoustic torpedoes: Build a microphone that can detect a nearby robot in autonomous mode. If you're playing a defensive autonomous, then you should be able to shut down anything that makes noise on your robot, while the attacking robot will be making lots of noise: driving, shooting, scoring noises. The microphone setup could be very simple: a microphone inside a cone atop a servo. Come to think of it, I don't think there is any way other than this to detect another robot, and I don't think I've ever seen it in action. Code would be very similar to the camera code. To make this a more challenging build, make a parabolic cone so that it detects only in a straight line, and soundproof it so that it still works when your robot is moving.
Edit: This would probably take quite long to do. You'd have to test several mics and tweak your cone so that it only detected things within 5ft of you. You don't want it to detect the audience screaming (and once your opponents determine that you have an acoustically-hunting autonomous mode, they'll be screaming). You'd also have to do lots of tweaking to make it as small as possible while still being useful. You don't want a satellite dish on your robot.

Idea #6
Wire-guided autonomous mod... hahah, just kidding

[Jeremiah Johnson]

Originally Posted by danshaffer

any chance we could get some pics?

Here's two attached photos. The first is from 2003 when we were painting it and the second is from the weekend before ship when we were testing it at the mall. It's used as a table for the controls more than it's actually used as a cart during build.

[Mike]

Originally Posted by Bongle

Drop a RC, OI, and the gyro/accelerometer chips in your programmers lap and tell him to do it (that's all he needs. He can tape the chips to the RC and just move it around to test).

That easy, huh?

Ahahahahahahaha

[team 1094]

in our off season we are doing alot of a appearances through out summer, and we are planning on going to another compittion. we also are doing mini learn session for the next season.

[Bongle]

Originally Posted by Mike

That easy, huh?

Ahahahahahahaha

That's really all the materials he needs for a basic one. The coding portion will probably be difficult, but he shouldn't physically need anything more to make a low-speed reckoning system. All I meant is that he doesn't need a working robot or lots of space to test this basic one. But nobody cares about a programmer's pain, so to a non-programmer team-leader, that IS how easy it is. You drop it off, and come back a month later and hopefully he did it.

Just realized I totally forgot about wheel-turn counters. You can use them too (in fact, they're much better since they aren't measuring the 2nd derivative of your location). But you'd still only need a rolling robot chassis, not a powered robot or lots of space for that.

Copied Idea #8:
Make a home-made wii controller and hook it to the OI like 1846 did at GTR. I'm not sure how controllable it is/was compared to the normal joysticks, but hey: technology for technology's sake.

Ridiculously hard idea #9 (requires medium-difficulty idea #5):
Once you've made your parabolic acoustic-homing microphone and it works, have its analog results stream in via the ADC and implement some fast fourier transforms on a buffer of that data. Once you've done that, experimentally determine power spectrums for crowd noise and remove. Once you've done that, have your scouts record the power spectra for various robots and upload them to your robot so that it can distinguish between a robot sent to interfere with it against the robot it is actually targeting for a hit.