Climbing Mechanism Ideas

[ToddF]

Originally Posted by TRIron95

Extending the arm over 54" seems to be in violation with the rule that says your robot must never extend outside a 54" diameter cylinder.

The max envelope is a cylinder 54" in diameter, 84 inches tall.

[engunneer]

For any teams that have a 3D printer, I just published a design that might help you think about climbing. I made the model for my desk at home so I can stare at it and imagine all the robots climbing up.

http://www.thingiverse.com/thing:40727

[mrnoble]

Originally Posted by engunneer

For any teams that have a 3D printer, I just published a design that might help you think about climbing. I made the model for my desk at home so I can stare at it and imagine all the robots climbing up.

http://www.thingiverse.com/thing:40727

That is super cool.

[ToddF]

Here's another cooperation idea.

The trickiest part of the climb is leaving zone 0 before touching the zone 2 bar. If you could sit on the floor and reach up to the zone 2 bar (touching zone 1 bar first, of course), inch worming up to the top isn't too hard.

So, robot A deploys a tarp from a roll and tows the tarp under the lowest bar, spreading it out to cover the ground. The tarp could be pretty big and not violate the 54" diameter rule. Robot B drives onto the tarp, thus leaving contact with zone 0, and inchworms up the pyramid, from bar to bar.

Robot A then reels in the tarp and hangs for ten points. Much less coordination between the two robots.

[Djur]

On the subject of coopertition...

A few kids on my team had the idea of a skyhook that would reach down and grab a friendly robot by attaching to the eyelets/connecting points that are meant for the belay system. Making a skyhook, though, will be tough.

[nathan_hui]

Originally Posted by Djur

On the subject of coopertition...

A few kids on my team had the idea of a skyhook that would reach down and grab a friendly robot by attaching to the eyelets/connecting points that are meant for the belay system. Making a skyhook, though, will be tough.

I thought skyhooks were in the same genre as snipe hunts. Odd that they would be useful for design ideas...

[SM987]

Just prop yourself up on an air cylinder "foot" on top of a frisbee, throw a batman hook to rung 2 (making sure to somehow touch rung 1 in the process) and hang for a quick 20, or climb for a slower 30.

[pfreivald]

Originally Posted by Randomness

It also depends on how you interpret the climbing rules... but it is my belief that robots have successfully climbed if they get to level 3 height without touching any parts of the pyramid.

That seems to be in direct conflict with the rules.

[Randomness]

Originally Posted by pfreivald

That seems to be in direct conflict with the rules.

So I wouldn't be surprised to see my interpretation disallowed on Q and A, but I do think that it works as the rules stand.

Here it goes:
"Points are awarded for the highest Level achieved for every ALLIANCE ROBOT that CLIMBS its PYRAMID. The Level to which a ROBOT has CLIMBED is determined by the lowest point of the ROBOT (in relation to the FIELD). CLIMB point values and Levels are defined in Figure 3-4.

A ROBOT has CLIMBED its PYRAMID if it contacts the PYRAMID in sequential order (Level 0, 1, 2, 3) during ascent and no more than two (2) Levels simultaneously.

If a CLIMB is considered unacceptable (e.g. a ROBOT has touched non-adjacent Levels or more than two (2) Levels at a time), a Referee will indicate a rejected CLIMB by turning the offending ROBOT’S PLAYER STATION LED strings yellow. The ROBOT will be ineligible for CLIMB points unless and until it begins a new CLIMB from the floor, Level 0."

So a robot that is boosted past 60'' is at the highest Level, and the rules say that a robot gets credit for the highest Level (the top) for a successful CLIMB (passing the following criteria). The question is whether or not it has CLIMBED legally. Did the robot contact the pyramid in sequential order? Well, probably - it certainly wasn't out of order, as all it touched was level 0, and the rules never specifically state that robots much contact each level that they want credit for. And the robot clearly hasn't contacted more than two levels simultaneously.

If you look at the (e.g ...) section in the third paragraph, then a boosted robot has succeeded even more easily - it definitely didn't touch non-adjacent levels or more than two at a time. That said, other scenarios that pass the (e.g. ...) part fail the first part.

Bottom line: would I be surprised if Q and A said that I was wrong? Not really. But it does make sense to make it slightly easier for teams to help others, as there are additional coordination/strategy problems with assisting others.

[pmangels17]

It seems that many people want to climb the corner. However, it is going to be near impossible to keep from spinning. There would need to be an incredibly complex system to prevent spinning without breaking the device

[Cal578]

Originally Posted by pmangels17

It seems that many people want to climb the corner. However, it is going to be near impossible to keep from spinning. There would need to be an incredibly complex system to prevent spinning without breaking the device

A robot climbing the corner, particularly from the outside, could use the floor and/or rungs to avoid spinning. It's not easy, but I wouldn't characterize it as "near impossible" or "incredibly complex". Our team is working on a few concepts along these lines, but nothing solid yet.

[JamesCH95]

Originally Posted by MetalJacket

That's one of the most complicated elements of this climb. It cannot be accomplished using a method that is intuitive to how a human would complete the task (aside from jumping up the levels but I don't think most people will feel comfortable with a 100-something lb robot jumping around 5'-8' in the air).

I think this is a bad assumption. Humans employ several different ways of climbing depending on the person, their build, and any athletic training they might have.

Originally Posted by pmangels17

It seems that many people want to climb the corner. However, it is going to be near impossible to keep from spinning. There would need to be an incredibly complex system to prevent spinning without breaking the device

I think that this is a bad assumption also.

[Alex.q]

It seems like a lot of ideas require claws or grippers which do not rotate on the pipe. Would a pipe wrench mechanism work in this situation? It would allow you to pivot your robot up in one direction without rotating back due to your center of gravity. I don't know if it would properly grip the pyramid or if it would damage the pipes, but perhaps someone could test this. (I say someone and not me because I am just an alum and my team doesn't want to climb past lvl 1).

[RSaunders]

Originally Posted by Alex.q

It seems like a lot of ideas require claws or grippers which do not rotate on the pipe. Would a pipe wrench mechanism work in this situation? It would allow you to pivot your robot up in one direction without rotating back due to your center of gravity. I don't know if it would properly grip the pyramid or if it would damage the pipes, but perhaps someone could test this.

After working on assembling a practice tower, I don't thing that a non-slip grip on a pipe is a good idea. The pipes we used weren't powder coated, and pipe wrenches visibly damaged them with less than 50 pounds of force. To grip the pipe without damaging the finish seems very difficult. I guess it's a function of how many square inches of contact patch you have, but the sub-inch contact patch of a pipe wrench isn't enough.

We've been looking at leverage against the tower support posts below the horizontal pipes, they are in the same zone. A pretty compact "claw" can make solid three-point contact with a single Level of the pyramid and support 150 pounds with compression contact only.

[cadandcookies]

Originally Posted by Alex.q

(I say someone and not me because I am just an alum and my team doesn't want to climb past lvl 1).

Hey! Not precisely true! We're prototyping for both level one and level three designs!

On another note, perhaps a dual engagement version of 2056's 2010 climbing mechanism would work. The first "climber" contacts the module, and the robot is lifted to vertical (or maybe so that the side of the robot is roughly parallel to the side pole, rolls up to the first bump, at which point the second set engages above the bump and the first stage disengages. Repeating this process would lead to a "hand-over-hand" approach to climbing the side.

Of course, balance would be a possible issue here...

Challenges like this are what makes FIRST so fun!