2363 Triple Helix "pre-release" video of Genome Eta
 

So, 4-5 inches of snow fell the night before bag day, canceling school. A small band of mentors and students braved the hazardous roads and fair weather drivers to spend one last day of testing before putting the robot in the bag. The parking lot to the school was unplowed, so some of us parked at a nearby park and hiked to the school, carrying gear on our backs. Since school was closed, we had no access to our normal test space in the auditorium. We had to make due with a tiny carpeted foyer. Our skeleton crew didn't include any members of the multimedia team, so I had to fill in. They will be putting out a more polished release video, but here is a quick and dirty "pre-release" video of Genome Eta.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

Amazing robot with lots of capabilities! We're looking forward to playing with you in Virginia this year!

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

Curious about how you ended up designing the recycling container grabbing arm. Why opt for the "passive" chain rotation of the second joint over, say, a 4-bar linkage? Chain failure seems like a higher risk than linkage failure, to me.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

I love seeing this robot completed. When I first saw your feeder station loading video weeks ago, I was amazed. And the systems you've added are very nice looking. I love the nifty design that picks up the RC.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

Love the way you guys grab bins looks effective and more importantly really really cool. Also did not realize anyone else listened to Pendulum.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

It seems to me that their design would be lighter, more compact, and simpler than a 4-bar linkage of equal length and strength, but I may be wrong.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

Awesome job as always! can't wait to see it in action @ VCU.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

Given the range of motion required, you're almost certainly correct about the more compact aspect. However, with the larger size and transport configuration rules, compactness is less important than most years.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

Love the RC holder mechanism. Super sleek and elegant.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

Like this one?


We tossed around the pros and cons of the two concepts. There were a few main points that made us lean towards using chain as a virtual 4-bar:
- That design plays to our fabrication strengths. The central tube is the only fabricated item, and the only precision dimension on it is the center to center distance between the holes. Minimizing fabricated parts count was big for us this year, as most of our mechanical team are rookies. For the 4 bar, we'd need to make 4 tubes and 8 end plugs for each arm. Before we ran out of money and weight budget, we planned to have two arms. Throw in a practice bot, and you're looking at 48 fabricated parts. For the chain arm, that drops to 4.
- Mechanical play. In my experience, for linkages to work well, they need to have bearings with absolutely zero play. In this case, that would probably mean good quality rod ends. For 4 arms that's 48 rod ends. Expensive.
- Actuation. Although the prototype had pneumatic actuation, we suspected you'd need more complex control than a bang-bang cylinder to do the things we want to do with the arm. Actuating the chain style arm is as simple as connecting a COTS gearbox to the tube. Driving the 4 bar arm isn't nearly as straightforward.

The nice thing is that the arm ended up with a weight of about 14 pounds. If we come up with a better way (cheaper and lighter), we can swap it out at our first regional.

Well, that's simple engineering design. If the tension on the chain exceeds its rated strength it may fail. The links, on the other hand, have lots of failure modes, tensile stress, buckling, thread pullout, etc. Used properly, both links and chain will work fine. Used improperly, they will both fail. As I say to the students, "Engineers don't need to guess. You can run the numbers and KNOW. That's what all that math you guys are learning in class is actually used for in the real world."

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

Thanks, Alex. We call it the Doc Ock Arm.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

Pendulum rocks.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

That's exactly the type of response I was hoping for when I made the inquiry. :)

However, with regards to the failure, I was more concerned with the chain potentially jumping/skipping or external forces on the chain than actually overloading the chain.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

I'm loving the design. Just curious, how much lining up is required to get your arm in the recycling bin, and how long does it take? Sweet robot, I look forward to seeing it work its' magic at comp.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

Gotcha. That doesn't seem to be a problem. When lowered very quickly against an immovable object, the arm support structure disintegrates before the upper arm chain skips. Unfortunately.

Myself, I prefer a nice mechanical fuse, rather than destroying structure.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

That is the coolest can manipulator I've seen so far. Another awesome robot from you guys. I really hope I get to take a closer look at it in St. Louis!

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

As long as the center of the internal gripper is within the circumference of the hole in the top of the can, it goes right in. Either the arm flexes or the can moves to accommodate.

That's a darn good question. What we know is that if the can is set up against a tote so the tangent point of the can is lined up with the hole in the handle of the tote, and the robot is lined up with the tote (think autonomous) the arm stabs the can perfectly. We also know that our vision is good enough that we can drive down a line of yellow totes and stack them. Testing with our practice bot will tell us if the vision is good enough to let us stack the yellow totes and also pick up the cans as we go down the line. If not, we'll think of something to make it work. There is always a technical solution to a problem. The trick is to find it and implement it elegantly.

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

Great robot!!! See ya at Chesapeake!

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

Absolutely loving this! I expect this to definitely be the best 2363 robot to date!

Way to just keep iterating and adding functionality... love the arms!

Re: 2363 Triple Helix "pre-release" video of Genome Eta
 

WOW! I LOVE that container claw! Very unique design! Great Work!