Hypothetical Situation - Stairs

[apalrd]

Or a worm gear. That would probably provide all of the reduction you would need.

I would probably mix a pair of encoders, a gyro (maybe) and an inclinometer. The encoders would determine wheel speed and distance, the gyro would (as usual) determine rotational velocity (Although you might have to do some stuff to keep it sensing correctly, or just use it when in a known orientation), and the inclinometer would detect the ground. An accelerometer *might* work but would be affected by the robot's forward velocity. A weight on a pot would work as well, but would have weight to it (and weight is always a bad thing).


David - about servos - They are weak motors with built-in logic to steer to a certain position, and work the same as a motor, Victor, and pot. Except they are weak. With a CIM or Fisher-Price as the motor, you have significantly more power, which with gearing means you can drive whatever load you must drive faster. Any motor can move the load, but a more powerful (not necessarily a higher/lower free speed or stall torque, but output power, measured in Watts) motor can perform the task faster.

[lynca]

Originally Posted by Dave Scheck

Check out our Encore promo video from that year to see more details.

Again, that was a different challenge with only a single step to deal with.

This is in my top 5 FRC Promo videos, I show this to students everytime they even think about making a promo video !

[Karthik]

Originally Posted by BrendanB

Did you have any issues with the tank treads coming off while in use or have suggestions that you guys used? I noticed that is was a common issue for team 195 and during a few drive train lessons have heard that treads have problem staying on track. Your 1114, 1503, and 1680's robots are some of the top tread robots in my honest opinion.

We had issues beyond belief. We were throwing treads all over the place. Most of the time we were snapping them rather than just having them fall off. This post discusses the issues a bit:

http://www.chiefdelphi.com/forums/sh...94&postcount=5

My the end of the season we had the belt trapped so it couldn't find a way to slide off. Unfortunately this really cut down our efficiency so our robot was slower than we hoped. Definitely not an ideal situation.

[Jay TenBrink]

In 2004 we happened upon a simple, robust, and fast method for climbing the stair. We put drop down ramps made of UHMW plastic on the front of the robot and had 4 large knobby pneumatic tires. The ramps would start the match folded up and would drop down and latch into place, giving the front face of the robot a sloped leading edge that would be just above the top of the step, sort of like an amphibious car.

To climb the stair the driver would simple turn the robot to face the stair and apply throttle. The ramps would lift the nose and when the front tire hit the stair’s edge it would climb onto it. Because the robot would high center between the wheels, we put UHMW plastic strips on the frame so we could glide across until the rear wheels would engage. To get down from the platform we would just drive off and bounce a bit. It was not a finesse move and required the robot to tip about 30-45 degrees, so you needed a relatively low CG. I don’t think we ever tried, but I think we could have climbed multiple stairs if there was a good edge to bite into like in 2004.

This method served us very well and was very KISS, perhaps the KISSiest

Jay

[rsisk]

Originally Posted by Cyberphil

The most interesting thing I found was something like this:

http://www.youtube.com/watch?v=uPvj-2KC9G4

But that would be pretty out of the question for FIRST.

Why would this be out of the question for FIRST?

[Jared Russell]

Surprised nobody has mentioned Whegs yet They are crazy cool, extremely simple, and can climb obstacles as high as the diameter of the wheel.

http://www.youtube.com/watch?v=F4GF2UFhv8Y
http://www.youtube.com/watch?v=P7UkXh8-cBw

And if you don't want to bump around on flat ground...
http://www.youtube.com/watch?v=lrE8CcquOuQ

[Cyberphil]

Originally Posted by rsisk

Why would this be out of the question for FIRST?

By out of the question, I did not mean that FIRST teams could not do this. I am sure that FRC teams could do this actually!

What I meant was the usefulness of this. The only thing this (What I guess you would call a drive train) is made for is climbing stairs. This could potentially be used on a bot, but the usefulness of it is what I would question.

Now after saying this, I know the 2011 game is going to consist predominately of stairs, and some team will come up with this design and make it to Einstein.

That would be just my luck, and I would love every second of it!

[Jared Russell]

Originally Posted by Cyberphil

What I meant was the usefulness of this. The only thing this (What I guess you would call a drive train) is made for is climbing stairs. This could potentially be used on a bot, but the usefulness of it is what I would question.

Actually, if properly configured, this design allows for omnidirectional movement (similar to mecanum drive) as well (http://www.youtube.com/watch?v=1uynm...ature=related).

[Cyberphil]

Originally Posted by Jared341

Actually, if properly configured, this design allows for omnidirectional movement (similar to mecanum drive) as well (http://www.youtube.com/watch?v=1uynm...ature=related).

Yes, it very well could. Like I said in my post:

Now after saying this, I know the 2011 game is going to consist predominately of stairs, and some team will come up with this design and make it to Einstein.

That would be just my luck, and I would love every second of it!

I would personally love to see this design and have it work! I would shake whoever's hand who decided to create it for a FRC bot.

[Hawiian Cadder]

i explored the idea of using a screw drive with first, the main problem is a lack of power, unless the screws were made of rollers, like a mechanum wheel, they require a tremendous amount of torque to move. i looked at ice fishing augers as a supplier for the screws, they are long and fairly perfect.

[BJC]

Originally Posted by Hawiian Cadder

i explored the idea of using a screw drive with first, the main problem is a lack of power, unless the screws were made of rollers, like a mechanum wheel, they require a tremendous amount of torque to move. i looked at ice fishing augers as a supplier for the screws, they are long and fairly perfect.

I guess if you needed ALOT of power you could make a single gearbox that runs all 5 CIMs (think 118 crab syle) and use pneumatics to shift the direction the auger is spinning. (as opposed to a high and low gear)

You could also significantly decrease the amount of torque needed to move by making the auger a slightly smaller diameter on either end so it's only contacting the floor in the middle.(think 8 wheel drive; the outer wheels are raised for the same purpose.)

[apalrd]

With an auger drive, how would you turn? You can't run one side forward/one side backwards, that would make it drive straight. You can't run both one way, that would make it strafe.

I don't think it would take as much torque as you think, since you are transferring the power to the floor approximately the same as you would in a mecanum drive. That said, you could always go with 2 CIM 1 FP per side (and have more power than 5 CIMs total) and leave a single CIM for whatever mechanism you want to build.

[BJC]

I think the only way to do it would be to run it as if you were going forward but with one auger going more slowly. It certainly would be an arc-like turn though.

All in all, in my opinion even if there was stairs in this years game augers would not be an eligant solution. They are simply over complicating the problem.

[Tom Ore]

This may have been posted before, but here it is again...

http://www.youtube.com/watch?v=nVzj3_MUBjE

[Hawiian Cadder]

i cannot remember, but wasent there a team that made a 12 wheeled robot for aim high, i believe they used roller blade wheels or something, nobody thought their robot would be able to climb anything but it was one of the best.