Fastest FRC Robot?

[Caleb Sykes]

Originally Posted by donkehote

It was surprisingly easy to drive. It had a double stacked omni on the front to turn. We drove forward on the small wheels, and turned when we rotated the wheel. we originally had Ackerman steering, but it never returned to center 100%. Using the omni, straight was straight. The rear axle was solid, but turning lifted the inside wheel, just like in a go-kart, so drag losses while turning were limited. It was a lot of fun to drive.

Did you incorporate any sensor feedback (e.g. gyros, encoders) to help the driver?

[donkehote]

Originally Posted by inkling16

Did you incorporate any sensor feedback (e.g. gyros, encoders) to help the driver?

We only had an array of ping sensors looking forward, and to the sides for auto.
it drove laps autonomously in auto with obstructions, but again, it never made it to comp, so I never got to see how well that would have worked in real competition.

[Michael Hill]

Originally Posted by RoboChair

There is Geared and Actual FPS. We only could hit 17-18 FPS on field(we measured this). Our 22 FPS was our theoretical top speed.

I vote that this thread reports Actual FPS as measured on the field, not what the gearbox was "capable" of.

I've always been annoyed when people say "Geared for [XX]". Is that according to the JVN spreadsheet? or is that actually measured? Simulated another way?

[Ether]

Originally Posted by Michael Hill

Is that according to the JVN spreadsheet?

And if so, what value did you use for the "speed loss constant" fudge factor?

[AdamHeard]

Originally Posted by Michael Hill

I've always been annoyed when people say "Geared for [XX]". Is that according to the JVN spreadsheet? or is that actually measured? Simulated another way?

When teams are comparing theoretical speeds, I think it makes sense to sense to use "geared for" assuming that means free speed. This way frictional estimates don't affect the number. This is essentially the rollout ratio times a CIMs free speed.

Any other number I think is only valid if actually measured, or from a team that has measured several of their similar drives and has decent enough friction parameters to estimate future drives.

[Ether]

Originally Posted by AdamHeard

Any other number I think is only valid if actually measured, or from a team that has measured several of their similar drives and has decent enough friction parameters to estimate future drives.

Quoted for truth.

[Lil' Lavery]

Without any standardized benchmarks to test completed drivetrains, I have no problem with "geared to" comparisons. It's still the most useful shorthand to compare different drive speeds.

[Michael Hill]

Originally Posted by Lil' Lavery

Without any standardized benchmarks to test completed drivetrains, I have no problem with "geared to" comparisons. It's still the most useful shorthand to compare different drive speeds.

The problem is robots that are "Geared to [x]" will not even theoretically reach that speed. If I say I geared my robot to 25 ft/s, it will never reach close to that. In fact, if I say I geared my robot to 20 ft/s, it will actually go faster than the one "geared" to 25 ft/s. A more useful metric would be something like "it goes covers 20 feet in 2 seconds from a stop". That's something that's actually somewhat easily measurable and useful.

[Chris is me]

Originally Posted by Michael Hill

The problem is robots that are "Geared to [x]" will not even theoretically reach that speed. If I say I geared my robot to 25 ft/s, it will never reach close to that. In fact, if I say I geared my robot to 20 ft/s, it will actually go faster than the one "geared" to 25 ft/s. A more useful metric would be something like "it goes covers 20 feet in 2 seconds from a stop". That's something that's actually somewhat easily measurable and useful.

The value of everyone using free speed to talk about gearing isn't in being able to literally use the number as is to know the robot's top speed. The problem with using adjusted speed is that everyone uses different constants and parameters to define what their adjusted free speed is. Geared free speed puts everyone on the same page, and the reader can apply their own constants to adjust that free speed if they so choose. It's just a known benchmark that's easiest for everyone to agree on in order to compare robots to each other.

[Mr. Lim]

Fastest means something quite different on our team:

How long does it take to travel X ft from a standing stop?

Where X depends on the game as each one has a different typical "sprint" length.

I don't think we've geared for a specific top speed in a long time.

As a result, we probably have some of the "slowest" robots in FRC...

[ErvinI]

Originally Posted by Mr. Lim

Fastest means something quite different on our team:

How long does it take to travel X ft from a standing stop?

Where X depends on the game as each one has a different typical "sprint" length.

I don't think we've geared for a specific top speed in a long time.

As a result, we probably have some of the "slowest" robots in FRC...

To be fair, if all teams reported this measurement, then all robots would be the "fastest" robots, however they would all be the "fastest" for different sprint lengths.

That being said, I would be surprised if the highest top speed of a robot was not one of the sprint bots of '08. Those little fellas looked nigh-uncontrollable.

[AdamHeard]

Originally Posted by Michael Hill

The problem is robots that are "Geared to [x]" will not even theoretically reach that speed. If I say I geared my robot to 25 ft/s, it will never reach close to that. In fact, if I say I geared my robot to 20 ft/s, it will actually go faster than the one "geared" to 25 ft/s. A more useful metric would be something like "it goes covers 20 feet in 2 seconds from a stop". That's something that's actually somewhat easily measurable and useful.

No one disagrees with the point you're making about robots not hitting their free speed.

The point others are making is that in absence of good data on friction, teams should report free speed as that is a more clear comparison to others.

TLDR; I'd rather have someones idealized (frictionless) numbers than their BS guess at friction.