On February 21st the Robowranglers unveiled our 2019 robot: “OVERHANG”. It will be competing at the “New Mexico” (Amarillo) District, Dallas District, World Championship, and maybe at the “Texas-and-New-Mexico-But-Not-You-Oklahoma” District Championship.
(Yes, we are double qualified for the World Championship, but we still don’t know if we’ll be going to District Champs.)
In 2018, we described Uppercut as our “Blackest Robot, Fastest Robot”. How does OVERHANG compare?
It’s definitely blacker, it’s not faster… but it does have more tricks up its sleeve.
After the 2018 Championships last year, we had lots of internal discussions about our robot design philosophy. Specifically: “If we could do it again, would we build another lightweight attack robot like Uppercut, or should we build Poofy-the-autobot?”
It’s a fun discussion, obviously 254 was far better than 148 last season, but we did enjoy those things our little black blur could do which their blue aircraft-carrier could not.
This year, we didn’t really build either of the two options…
As with everything in life, Overhang required making lots of tradeoffs as we continually try to find the optimized balance between our LIMITS and WANTS.
Size / Weight
Overhang is small. You can’t really see from the video, but it’s not nearly full size. We’re expecting to face lots of defense in this game. (There seems to be something about a black robot which makes it a magnet for kitbot drivetrains.) In particular, Overhang is narrow, and we’re hopping it can shoot through gaps in traffic.
Overhang is light… not “Uppercut light”, but still not nearly max weight.
Okay, honestly… we think Overhang is a bit fat by our (unreasonable) standards.
We’re really disappointed by how heavy the robot was at the end. We put those dang pneumatics on the robot and the next thing you know, every designer on the team wants to put another widget on there. It adds up!
Does the robot need to go on a diet? No.
Will we lighten things up as we iterate them? Of course.
At the beginning of the season I really wanted an H-drive similar to X019. We put a prototype disc scoring mechanism on X019 and it was SOOOO fun to watch on the field.
But… we’re concerned about how defense will affect gameplay. There are some big bottlenecks on the field. There is a large penalty zone. In one minute I’ll be thinking about how an H-drive with omni-wheels will slide off defenders and roll their way to one of the MANY goals on the field. In the next minute I’ll be thinking about getting pushed deeper and deeper into the “alley” until we’re forced to cross onto the far side of the field for a penalty.
We chickened out. We built a 6WD, and have been playing with the wheel configs since then. At the time of the unveil video, it had omni-directional wheels on the outside and traction wheels in the middle.
How does it handle? Very smooth. While I’m partial to the “J-turn” wheel configuration, the drivers seem to like this setup. It jumps off the level 2 platform no problem, climbs over the steel barrier into the depot, and drives up the ramp from all angles (yes, it even drives straight up that dang corner).
We are using (3x) MiniCIMs, single-speed. Geared a little slower than last season (since we’ll probably never be doing “full field” runs). This year we did a new powertrain configuration nicknamed “stilt drive” which allows for high ground-clearance while using 4” diameter wheels.
We chose NOT to score off either side of the robot. At one point we were considering a turret. At one point we were considering a pass-thru. At one point we had the disc scoring on the back and the ball scoring on the front. (Concept Name: “Mullet”).
We emphasized “after the buzzer” climbing to platform level 3. This required integration of gas-shocks and made that mechanism more complicated (but sooooo much cooler). It’s fast. We can get to the wall with 1-second on the clock and still make the climb.
We said the “home” position for the elevator at the bottom hard-stop must have the disc scoring beak be at the correct height for the loading station and low goals.
We wanted to be able to pickup objects and score them in all the “low level” goals without moving the elevator.
We wanted the disc / ball elevator presets to be the same for both the middle & high rocket goals.
We (eventually) gave up being able to pickup balls and discs without extending beyond our frame perimeter. This means: we can’t pickup game objects on the opponent’s side of the field (if we ever go over there).
We (eventually) gave up on our floor disc pickup.
We wanted a “full bumper” all the way around the robot without gaps for intake or end-game. Breaking the bumper would have made some mechanisms much easier.
We decided everything MUST be in full driver-camera view. So if we had something scoring “off the back” we’d need to flip the camera around or add a 2nd camera (or a mirror… we talked about a mirror).
We emphasized FAST disc scoring and acquisition from the loading station as our primary function. We agree with many “pundits” who say scoring discs is going to be harder than most people think. Lots of other systems had trade-offs to accommodate this subsystem. We’re very proud of how “forgiving” the mechanism is to misalignment caused by defense (or bad driving). The best time to help the drivers look amazing is during system design!
We added pneumatics to the robot. This is not an easy decision for 148 to make. The “beak” subsystem for discs REALLY seemed happier with pneumatics, and we decided to take the plunge…
We decided NOT to do a partner lift for the endgame. We want to be the “last robot to climb” and score until the very end of the match.
We wanted to be able to handle balls of all sizes: +/- 0.5” is a big tolerance range!
We (as always) wanted to be as lightweight as possible, but gave up “pieces of lightness” to achieve many of the items listed above.
We wanted high ground clearance to get over the ramp in all configurations and over the steel bar in the Depot.
We wanted 4” wheels to enable a “longer” support polygon and we wanted to make the items lifted by the elevator as lightweight as possible: making the robot more stable when “up in the air”.
How’d we do?
I’d say, we get a C+ grade.
While some would say “C’s get degrees” I’d say “D is for diploma!”
I mean, I’d say: “We still have lots of time to improve, we’re barely at the midterm!”
Do I like this robot?
Oh yes. This one is a fighter. It’s got lots of little things which make it great. I think it’s already a winning machine, but I know we can push to make it better (hence the low grade). I see the potential of this machine, and I’m proud that we’re already pushing to take it to the next level.
Robots Switched at Birth
Have you seen Flyby?
Flyby looks fantastic!
(Fan boy moment: That PTO mechanism with the lexan gears at the top of the end-game mechanism? So beautiful!)
We’ve been talking all season about how Overhang is more Robonaut than Robowrangler. After seeing Flyby I’m genuinely convinced our robots got switched at the paint-shop. You paint that sucker in black, and people would swear it was built in Greenville. They’ve even got my favorite ever mechanism, the rocker drive!
Are We Dumb?
This is Fine.