Arm or Elevator - Adv + Disadvantage

Just curious to hear peoples opinions of the advantages of making an Arm vs an elevator to lift the cubes. Assuming a team can execute both correctly using motion profiles and closed loop feed back for consistent accurate control.

In general which would be more work? A cascading elevator with a cube manipulator or an arm with one main joint and a wrist for a cube manipulator?

Which would be harder to control. At first I though arm but after thinking about cascading elevators you always have stretch in the cables used to activate each section. So I am torn. Does anyone have experience with both and have a strong opinion one way or another? Curious to hear some thoughts?

I realize my questions could easily be answered with the “Depends” response. I get it, but people still have opinions regardless and I am interested in those. I realize the design has inherent variables that affect complexity and quality. Assume the final arm or elevator are equally effective.

It’s a lot easier to keep the heavy-duty motors and gearboxes low with the cascading elevator, which will reduce tippiness both when placing cubes and dashing across the field. Our current leading design is neither of the ones you listed, but also lets us keep the heavy stuff low: a scissor lift.

TL;DR my opinion favors elevators over arms

Considering the fact that scoring on the scale isn’t as high up relative to other “high goals” in past games (ie 2011) a cascade system may be a decent and compact choice over an arm on a pivot. With an arm, you’ve gotta account for the weight that’s going to be at the end of it, and at which height the pivot point will be (down low by the chassis, somewhere in the middle, or up at the max starting height?). With an elevator, you can put the lifting part at the very front of the robot and have a cube grabber/deployer as its own system on the lift. This way, You’re not trying to compensate for as much weight at the end of a lever. (254 in 2011 is the design that quickly comes to mind if you need to visualize what I’m talking about)

Just from our own experience with elevators and arms, more points of articulation created more points of failure.

We preferred the elevator but arms can work well is certain scenarios

Be careful with scissor lifts. They tend to be heavier than either an arm or elevator and require extremely tight manufacturing tolerances to work since each joint has it’s own tolerance stack up and is another possible point of failure. I’ve never seen a robot successfully implement a scissor lift.

Another thing to consider is elevators tend to be faster. If you try to move an arm at the speed of a really good elevator, there’s just too much inertia and it makes the robot unstable. See 469 in 2011 for what I mean about speed.

When debating elevator or arm the 1st thing I look at is “will I have to reach over other robots to score”? If the answer is yes then arm is your better route. If the answer if no I will lean towards elevator every time. I feel it is easier to build a fast and solid elevator than it is to do the same for an arm.

Over all its all preference and there really isn’t a wrong answer here as long as it does its intended purpose when finished.

typically an elevator may be more valuable but in this instance using an arm as a “swiss army knife” or all around tool could prove to be the better option. (climber, scale tool, switch tool, intake, outtake).

There is a 16" rule during matches except during end game.

Section 8 ROBOT Rules V067 of 133
R04.ROBOTS may not extend more than 16 in.(~40cm)
beyond their FRAME PERIMETER (see Figure 8-1), except during the ENDGAME
(see G05).

So you will not be able to reach over another robot to get cubes.

This matters, but perhaps not the same way you mean it-- remember that an arm is rotating about a point-- this means there is a point where it hits maximum distance outside the frame perimeter. It is very easy to design an arm that can reach the top of the scale but isn’t legal because it extends too far outside the frame perimeter. Also remember that the peak height you need to achieve is 6 feet plus change for the rim of the SCALE and potentially even more for trying to place on top of existing blocks.

Why not both? Have an arm on the end of a single stage elevator similar to 111 in 2011. You get the added simplicity of not having to make multiple stages of elevators, while being more compact than using a long arm.

Our team considered a scissor lift but quickly ruled it out after how much weight the materiel would add up to and that it might be too slow "depending on what driver you use. I’m interested to see how it works out for you guys.

From past experience, we have found that scissor lifts or scissor jacks are very good at expanding from a compact form to get a lot of linear motion, sort of. Because there are a lot of pivots, it is difficult to guide where the end of the lift goes. When we were successful with a scissor jack, it was only used to provide the push and we used other methods for guidance.

It would also be instructive to calculate the static forces and static torques at the pivot(s) and compare them to the ratings of the components you intend to use. Yes, this is a plug for applying some STEM education vs throwing mud against the wall.

The difficulty is that your robot is not static. When it moves, the arm will tend to bounce, causing shock forces and shock torques to be applied to the components. The shock forces and torques can be many times higher than the static ones but can be difficult to calculate. I think the ME’s at work typically estimate the shock forces/torques and apply a large margin to the components, then test extensively. If you don’t do this, you are likely to have the experience we had in 2016 where we blew up multiple VersaPlanetearies, disabling our arm, or twist hex shafts the way Spectrum did in 2015, leading them to totally rebuild their robot to change to a different concept on the practice day.

All great points and insight. Thanks for taking the time. We def have points to consider from this.

I don’t know if anyone else has pointed this out yet, but the geometry for a single point of rotation arm is pretty difficult to pull off with the dimensions given.

A single point of rotation arm probably requires a wrist, to be able to reach from the floor to place a cube on the scale.

That arm could be a 4 bar arm, if you want to keep the orientation of the wrist the same through the range of travel.

Someone mentioned that 5 of 6 robots on Einstein in 2011 had elevators. :wink:

In terms of elevators, what is everyone’s opinion between Cascade vs Continuous?

As I’ve been telling my student when discussing manipulators both before and after kick-off, it all depends on the geometry of your robot. Both arms and elevators can be executed well enought to accomplish just about any pick-and-place game challenge. Consider your design requirements and lay out the range of motions that will be required for your manipulator to accomplish those requirements. Consider the shape of your drivebase and the size of your end-effector. Consider the maximum size (both starting and playing configuration). Consider the multiple variations of both arms and elevators that exist.

There are some inherent physical differences between arms and elevators:
Arms tend to have longer lever arms around the power source than elevators, which is an important consideration when adding the weight of the game piece to the end of the arm.

Arms may naturally have the ability to reach beyond your frame perimeter as they rotate. This is important for acquiring game pieces outside your frame perimeter, but also comes into play when considering the allowed 16" extension beyond the frame perimeter per the rules.

Arms tend to shift the location of center-of-mass of the robot as they rotate.

Integration between elevators and end-effectors may be non-trivial, especially when factoring in your desired position to acquire game pieces.

Personally, I look for solutions that allow me to use a single-jointed arm first. Failing those, I look for solutions that allow me to use a single-stage elevator. If neither of those meet all of my design constraints, I then consider multi-jointed arms, multi-stage elevators, or some combination of the two mechanisms.

Who says you need to reach to the floor? Looking at the other bots on Einstein in 2011 should answer that question. :wink:

Continuous (generally) lifts the stages in the order of increasing mass. Cascade lifts all stages proportionally. There are benefits and drawbacks to both.

+1 to the speed of 469’s 2011 elevator. Here’s a match with some of the notables from that game:

Well… they updated the rule book… no more elevators. :frowning: