[Request] Robot speed rules of thumb

[Leav]

Hello,

I have been in FRC for a while, but for some reason never managed to learn good ball park numbers for reasonable robot speeds.

So let's use the wide open 2014 field as an extreme example, what would constitute a "sane high speed" for 2014?
at the other end of the spectrum, any idea of how to form a "lowest reasonable speed"?


According to the JVN calculator, our numbers for last year were:

• High speed 14.72 fps (18.17 fps theoretical)
• Low speed 4.00 fps (4.94 fps theoretical)

I wonder how these numbers fair in comparison to general wisdom about advisable speeds.
Right now we are shooting for ~11 fps for 2015, but i'm now thinking we might want to aim for a lower speed for increased precision.

Thanks for the help!


-Leav

[Jeffy]

Rules of thumb will differ from team to team, but here is what I have used:
7 ft/s Slow
10 ft/s average
13 ft/s Fast
>16 ft/s Might be too fast to control

As for "precision" of a drive train.
Many things other than speed really effects this. If you build a stiff, tight and low friction drive train you will have very similar precision positioning ability at many speeds. Also programming an exponential throttle response can help the driver with low speed interactions. Trouble can be had when the skid friction needed to turn is high. This can cause the driver to have less control of the exact position during turning
For this year, I'd suggest you just build a robot with very little skid friction.

Just remember that top speed isn't really whats important. It's the point-to-point time for a robot that is important.

[DonRotolo]

I essentially agree with Jeff, however I think slow is a little bit slower, and fast could be a little faster. I do agree 16+ fps is a challenge to control, but you absolutely must factor in how well your drivetrain is put together, since a good one is easier to drive. The quality of your driver, and the practice time they get, is perhaps the most important factor.

[IronicDeadBird]

Speed is an interesting factor this year considering per rules given there is no reason why a robot without a drive base couldn't score. If you are using these "rules" to determine what speed your robot should go I would suggest you consider otherwise because speed (in my mind) isn't a really important factor this year.
I'll try to make this simple but its hard for me to explain its just something that I observed nobody taught me this so I couldn't dig up a lesson plan.
In this kind of game "speed" isn't as important of a measurement as "idle time" you want all mechanisms working together to make sure your robot is always performing an action that would contribute to scoring. Wasted time is wasted scoring potential in that regards getting to a point on the field quickly isn't as ideal as getting to a point on the field when you are ready. If you zoom over to where you need to be but you aren't quite ready to off load a stack I would refer to the time spent fidgeting as "idle time". What sucks about "idle time" is that time isn't enough to increase your scoring potential. Its enough time that you lose scoring potential but not enough time that you can do anything meaningful to make up for it. The speed of your robot and the mechanisms on it need to work together to minimize this sort of idle time.
Hope that makes sense! If not I'll try and rephrase it as a Haiku...

[smclean1969]

The "right" speed is difficult to provide a universal answer for. In general, if your robot is an offensive based robot, it needs to be faster. The right speed should be fast enough that you're able to control both the robot as well as any field elements. We've traditionally had a top speed around 15 fps. For ultimate ascent we had a top speed of 12 fps with increased control. While the better control was nice, we were limited to two runs whereas the faster robots were getting three in. Last year, we had a top speed up around 18 fps and we routinely lost control of the ball because we were so fast in driving so we actually drove at speeds that were less than optimal.

For the low end, that's key for defense or for high control/high torque scenarios, such as the bridges in Rebound Rumble. I don't see the need for a low gear this year. For those defensive push matches 4 to 6 fps should be sufficient.

I agree that this year will be about deliberate control and effective pacing more than incredible speed. That said, the robots that will be top seeds at St Louis will be incredibly fast but they also don't need this blog for inputs. We all tend to follow their lead.