I’m working on a presentation about simplicity that would be similar to the presentations by Adam Heard. Along with this I would like to compile a list of simple robots over the years. I’m also thinking about simple robots relative to their performance. I’m also not worried about the drive train as I am mostly focused on the mechanisms.
Here are my primary examples from this year:
9098: simple robot that only scores cubes but is fast and needs very little resources
695: good all around robot that can pick cones and cubes from the floor and score anywhere. It only uses an elevator and everybot style intake.
973: compact, no pneumatics, fast, doesn’t need to get the game pieces in a gap.
2910’s Deep Space robot (2019) is one of my favourites that year, and IMO was an extremely clean and simple design. Not to mention an absolute beast on the field.
There is this presentation by Adam Heard called “Simple Robots That Win” that runs over a few good simple robots from over the years. The slides and videos are a bit hard to see because of the glare, but you can find the slide deck here.
Personally I think that 2910 is the greastest example of simple robots there is. Since 2018 they have built super simple robots that perform spectacularly at the highest levels of play.
Edit: IDK how I missed that you mentioned Adam’s presentation in your OP, oops. But I guess the links there now for anyone to quickly get to it.
Also sorry Brian, I didnt mean to reply to your post and I don’t know how to fix it.
Careful equating “no pneumatics” with simple, a lot of times linear motion is way more difficult to package and do cleanly (let alone light weight) with motors.
To add to the list of simple and effective:
148 in 2009
A whole host of 2011 machines that had simple extending tube arms
4656 in 2013 (low goal frisbee dump, not sure if match video or images survive)
I may be a little biased but 5727 Omegabyte’s bot this year was incredibly simple, the arm was aluminum extrusions and the intake was fairly simple, and no motor for the wrist, because theres a v4bar
973 generally has some pretty sweet simple robots. They often make me kick myself for not thinking of the obvious solution that has limited excess complexity but is highly competitive.
Here in New England again, 133 from every year. Their robots are always the simplest way to achieve the game tasks, with weeks of drive practice and great execution. This year their manipulator was a pinching claw (not touch it own it, no spinning wheels, only from double substation) and they were NE DCMP finalists. Every year I watch their robot and think “why didn’t we just do that…”
Adding to the list again, I’m surprised 2056 hasn’t been mentioned yet.
They are by all definitions a world class team, who consistently make simple/elegant machines and then push said designs to their absolute limits.
There is the aspect that they tend to do all the given tasks each year, which definitely increases the complexity of their machine from what it maybe could be if it “specialized” more for a specific game task; But I’d still consider them a phenomenal example of how to design and implement a simple robot.
The examples in this thread given as “simple” bother me a bit. I think of “simple” as something like one degree of freedom plus a spinning roller, or two degrees of freedom, but done in a way that simplifies the system’s controls (limited ranges of motion, one of the DOF is pneumatic, etc.).
Telescoping tubing mechanisms (greater than 1 stage, or any telescoping thing prior to the release of telescoping tubing COTS kits) are not particularly easy to build.
Cube shooters like 1561 and 1923 are excellent examples of simple top-tier robots this year.
Top tier everybot+ designs like 3176 and especially 1591 were excellent with limited degrees of freedom.
Slightly more complex:
5414 had a single degree of freedom arm + rollers for cones and a separate cube shooting mechanism (which they did make pretttty complex)
695 appears to be a 3 stage elevator, which is more complex to build than a single jointed arm, but they offset that by having no wrist or separate intake mechanism
With the availability of several COTS elevator kits now, I don’t fully agree with this, especially if everyone is using brushless motors (encoders for creating set points and PID control are way easier than in the past).
My recommendations (and im assuming some others here) were not based on being soley “simple”. They were based upon what OP asked:
Based on this, I stand by the teams I’ve listed and their “simplicity”.
“Simple” can also be rather subjective given the vast differences in manufacturing capabilities between teams. For a team with only access to hand drills and hacksaws, teams like 2056 or 2910 would likey be far too complex. For teams with access to CNC/precision machining, I’d absolutely consider those teams/designs as simple.
While its on-field performance could have been better, I can’t really say I would have wanted the overall design to be different all these years later. Sometimes I think about what could have happened if we had a better camera mount, the radio wiring was more stable, or the auto was more consistent…
