Pages

Wednesday, July 30, 2014

Words About the Rookie Project (WARP): Team name and ongoing work

Back again after a while, I'll be trying to update more frequently now that the ball is really rolling on this.

Plenty has been going on, with plenty of work being done by a few mentors, and the students that we are carrying over from my old team. The big news item is that we've selected our team name. We will be known officially as the Worcester Area Robotics Project, but will generally go by the acronym WARP. I like the name. It's succinct, memorable, and hits hard. I can easily see it being used in play-by-play dialog. It's also relatively unique, with only two really similar hits on the blue alliance: Warp 7 from Toronto, and the defunct WARP (Wasatch Academy Robotics Program) from Utah.

Some considerations we used in brainstorming and selecting a team name.

  • Must be relatively unique and identifiable. Due to the alliance structure, and the incredible bonds that can form between FIRST teams, it is important for your name to stand out and be memorable to your fellow competitors. We wanted to do something to distinguish ourselves from the crowd, rather than be yet another "Robo-(insert school mascot)." Teams like the Thunderchickens, Cheesy Poofs, and Exploding Bacon take this to another level, with names that border on the absurd, yet are instantly memorable. 
  • Must be usable in varied professional contexts. All FIRST teams have to do a lot of cold calling to companies for support, strive to become community fixtures, and present their work to numerous individuals and organizations unfamiliar with the program, and the half sporting event, half party atmosphere of a FIRST competition where it's all too easy to think about the team name being used. The name must work both in a competition context, and as an element of a professional letterhead, as a charitable event sponsor name, as a program supporting very young kids, at the top of a flier for prospective parents, and any other context that the team name might be used. We had to nix a name that was very popular with pretty much everyone within the team, and that worked extremely well from a competitive and branding perspective, because we were worried about how the name would be perceived in more professional contexts. 
  • Must not carry any side-meanings with unfortunate connotations. Be careful here, it's easy to pick a name that has elements, alternative uses, acronyms, etc. that carry distasteful meanings. Generation gaps in either direction can make the problem worse, as one group may not be aware of an alternate meaning. Make sure that all names are appropriately vetted. Urban Dictionary can be a good place to start. 
  • Must have branding potential. To be seriously considered, name suggestions had to come with ideas on how they could be incorporated into a cohesive team brand, whether it be through colors, logo concepts, robot name patterns, team uniform ideas, competition props, or other elements of a consistent team identity. 
  • Most importantly, the team must be excited about it! The team name should be something that members are happy to say that they are a part of, and proud to wear on a shirt. If your members won't embrace and promote the team brand, it won't succeed. 
We're still working on some of the other brand elements. We've narrowed down team color and logo options to a few finalists, and will be further developing each of them, aiming to make a final decision shortly. It's important to get this done early, so that the team can use its newly created brand to promote itself for both fundraising and for student recruitment. 

In other news, we have been in talks with the local 4-H extension office about organizing as a 4-H club. This is critical for us as a school-independent team because it will provide us with 501c3 status and liability insurance, two things that are typically taken care of by schools. 501c3 status means that we are recognized as a nonprofit organization, and exempts us from paying taxes on income. More importantly from a practical perspective, it means that donations to us are tax deductible, and verified to be going towards a good cause. This is a requirement for any serious fundraising, as most corporate grants and sponsorships are nearly impossible to get without being considered a 501c3 organization. Liability insurance will protect us financially if, god forbid, an accident happens, and is also a useful thing to demonstrate that you have in a few situations, such as any kind of rental. 4-H also offers a number of other benefits to us, including a large recruitment network, several outreach and fundraising opportunities, and takes care of many things such as medical forms and various protection policies that we can use rather than developing ourselves. Nothing is final yet, but talks with them have been encouraging. If you are interested in forming a team independent of a school, I strongly recommend giving 4-H a look. 

A lot of writing has been happening as well, including a team handbook, bits and pieces of a team website, and training materials for new students. I'll post more about these as they approach completion. 

200 days to bag day 2015


Tuesday, June 17, 2014

Words about the Rookie Project (WARP): Welcome

I found out recently that my former FRC team is not going to be able to continue for the 2015 season, due to circumstances at the school that hosts it. Team assets will be reclaimed, and students will loose a fantastic opportunity. Not cool.

Today, our journey forward starts. Myself and a few mentors from the team have regrouped and are looking into beginning anew as an independently operated team. Since this is a large undertaking which is quite daunting to me, I imagine it's pretty daunting to most everyone who has faced the prospect of starting a rookie FRC program. Therefore, I thought it'd be a good idea to document the process as extensively as I can. This blog will be part day-by-day log, part resource dump, part inspiring and interesting videos/articles, and part musings on the competitive robotics world, in an attempt to paint a picture of what goes into building an FRC team from scratch.

We will have a few two-year students from my former team joining us, but not enough to consider ourselves a veteran in FIRST's eyes. Thankfully, this will allow us to take advantage of the numerous resources out there tailored specifically towards Rookie teams. This is a bit of an annoyance to me, that there's such a gap in resources offered to Rookies vs. struggling veterans, but it is what it is. One goal of this blog is to help to extensively document the process of acquiring resources to field a successful FRC team, a skill that can help a veteran as much as any rookie.

Even with a few rookie funding opportunities at our disposal, we're going to face an uphill battle. We will be starting at essentially zero when it comes to tools, materials, documentation, and policies. We will have to take a completely new approach to student recruitment, since we no longer have a single school to feed us with students. We don't have a place to build right now, or an umbrella organization to provide us with non-profit status or accept funding on our behalf. While between ourselves, our mentor body is petty well versed in each individual aspect of FRC, we don't have much experience putting it all together and managing a team as a whole. It's going to be hard. But it wouldn't be fun otherwise, right?

245 days to 2015 bag day

Thursday, June 13, 2013

What vs. Why with Past Designs

When we design our robots, it's natural to look to the past for inspiration. Robots every year show up left and right with bits and pieces of champions from years past bolted on, because these great robots clearly did something well. This learning from the past pushes the sport of competitive robotics forwards, as teams build their robots upon each other's shoulders, and quality increases each year.

But sometimes, things can go a little far. Some challenges have game tasks very similar to other years, which can tempt designers to borrow not just bits and pieces, but whole designs. It's a good way to start things out. But be careful! Successful designs are subtle things, and games have subtle differences. Before duplicating a design, you should understand exactly why the design was successful in what it did. You need to get very specific. "The NZ design was good because it was fast and simple" is not a good way to look at things. It was fast, and it was simple for Gateway, but in the face of a new challenge, you need to reconsider things. You need to be asking questions like "does the new game have a gamepiece that justifies locking myself into a fixed width, narrow intake like the side roller design?" "Do I need to score out the end I pick up?" "Do I need to have the controlled ability to score a single gamepiece at a time?" "Do the field obstacles justify a more complex lift system?" Some of the answers will lead you to remove features from the old design, others will lead you to add new ones. Developing this deeper understanding of the games, past and present, will lead you to better designs, as you can now design to precisely match the game's requirements, rather than just tweaking something that worked well in the past.

If there's interest from the VRC folks, I can write up my comprehensive list of "Why the 6 bar, side roller design was great for Gateway, but is not a universal "efficiency design"," seeing as this design has been the starting point for so many robots over the past few years.

A related issue is teams that mirror a construction technique or style, like the Cheesy Poof's West Coast Drive, without understanding it fully. "The Cheesy Poofs do it that way" isn't good design justification. In order to take advantage of a design like this, you have to understand the rationale behind it, and the physics in play. Otherwise, you run the risk of not getting the full advantage of the design, or even worse, causing a failure, because you didn't understand how it works. Again with the West Coast Drive example, one of the primary reasons to cantilever wheels is the weight savings that comes with no outer support for your wheels. And yet I can't count the number of west coast drive designs I see bulked up beyond what they need to be, which end up heavier than the kitbot!

Always understand why a design you've been inspired by was successful. Not only will it give you a better understanding of the design you are thinking about borrowing from, it will give you experience breaking down a problem, which is the core of engineering.

Friday, April 20, 2012

First Thoughts on Sack Attack


  • The game should have been called Bridge Battle's Revenge.
  • I think the GDC named it Sack Attack to rid the robotics world of the memories of Stack Attack. 
  • I hope those sacks are durable, and their contents are easily cleaned up
  • If they are though, I think "sacks" are a brilliant gamepiece. Unorthodox yet doable, and doable in a vast number of ways. Good size too. They should "stack" pretty easily in robots. 
  • 98, non alliance specific, unorthodox game pieces, some of which are shiny and give extra points. Hello, year of continuous intake improvement. Excite!
  • Glad to see the anti-wallbot "trapping" rule, but I don't think they went far enough. The field has one heck of a choke point in the middle. 
  • My first thought is that allowing descoring, and making it so easy, is a huge mistake. Descoring quickly and efficiently in round up was an interesting engineering challenge. Descoring in this game is as simple as building RakeBot, or in the case of the high goal, StickBot.
  • A lot of inexperienced teams are going to try to build claw bots. While I'm sure an elite few will emerge, I don't know how claws will fair by and large in this game. 
  • We're going to see a lot of eight wheel drives and tank treads this year. People are going to discover in a hurry that the sacks are just as much an element of terrain as they are a game piece. If I was playing, I'd totally build my VEX tristar wheels just to try them out. 
  • The manual referring to Polycarb as "non shattering plastic." Perfection. Watch out, sacred pneumatic components, you have a challenger for favorite phrase in a robotics competition manual.
  • I have mixed thoughts about how human-autonomous interaction was brought back. On the one hand, I would have preferred to see all but the "forgot to turn on the robot" rule gone. But I'm glad that the game looks to be less influencable from the alliance squares. I'm also glad that programming skills interaction has been reduced, though I'm not yet sure the GDC went far enough there. 
  • This is going to be either a really good year, or a really bad year for autonomous. It's bad in that the trough is an easy target for an autonomous robot. It may be good if the dynamics of the sacks throw off drivetrains just enough to force them to use advanced corrective techniques. Of course, that might just make it bad again...
  • This is a game of volume. Volume pickup, and volume scoring. Large goals, relatively small gamepieces. 
  • Unless...The "game breaker" that jumps out at me would be a "super stacker" on the high goal, executed at the end of the match. 
  • The troughs are at the perfect height. High enough to be a challenge to score in, and not restrict robot design too much. But just low enough to force something very, very creative to reach the high goal, and cross under the troughs. 
  • The endgame seems to be designed to force a split second choice between going for it or a last second big play. I'm betting that 95% of the time in high level play, the last second big play will be the right move. 
  • I think this game will be exciting to watch. I think it will be easier than some recent games, though still nowhere near as easy as I'd like, to keep track of the score. 
  • However, I am afraid that offensive powerhouses are going to get shut down by descorers. Again, for a game to work with descoring, it must be substantially harder to descore than to score. That doesn't seem to be the case here. 
  • I don't see one, but I hope that someone finds a way to effectively "lock up" gamepieces, much like green eggs did in 2011. 
  • I like the small bot, big bot dynamic of the college variant. I wish there was more to encourage differences in their design though...the obvious strategy seems to be score, score more, and keep scoring, faster than they can descore you. 
Overall, I guess the minuses outweigh the pluses for me. But I'm sure tons of teams all around the world will have a blast proving my reservations wrong.

My hopes for the new VRC Game

The next VRC game is due to be released in just a few hours. Although I don't plan to be involved at all next year, I still like to see a good year of competition, and a good game. Gateway has largely been a disappointment to me, and I'll be writing quite a bit about this at the conclusion of the VEX world championship. But anyways, my list of things I hope to see:


  • A challenge where lifting objects to a great height is not the grand engineering challenge of the game! It's been a common theme in the past several years, and people are getting too good at it. I'm more than a little tired of parallel 6 bar lifts...
  • A real emphasis on drivetrain design. VRC has never been played on anything but a flat field, and the congested fields as of late has limited the effectiveness of "zippy" designs. 
  • On that topic. I doubt it'd ever happen, but I think that modern VRC robots have outgrown a 12x12 field. Six max dimension FRC robots with bumpers take up just over 3% of the floor area in their 27'x54' field. Four 18"x18" VEX robots take up 6.25% of their field, and most VRC games are much more relaxed on expanding robots. 
  • Get rid of, or drastically limit, autonomous-human interaction. This has taken much of the challenge out of autonomous, and made the gateway programming skills challenge an absolute joke. I want to see another 1103...
  • I don't mind human loading, but don't allow teams to rely completely on it. 
  • A gamepiece we've never seen. Footballs in clean sweep were great, and made "scoop" style pickups viable and optimal for the first time since the very early early days of FRC. 
  • An endgame. Rebound Rumble has reminded us all the power of an exciting endgame, and VRC needs to step up their game in this area. 
  • A clear winner. I watch robotics. I love robotics. I live robotics. I know the rules the day the game comes out. And yet, the past two years of VRC, I've only had a vague idea of who was winning the match. Gateway and Round Up pulled the audience's head in several directions at once, and demanded fast mental math of us. VRC needs real time scoring. Even pseudo-real time scoring systems like in Logomotion or Lunacy made a world of difference for the audience. Or, do a fantastic visual game like clean sweep again. 
  • Either overwhelm or underwhelm the robots with the size of the gamepiece, but don't go in between. I think two of VEX's best gamepieces come from its earlier days, in Quad Quandary and Hanging-A-Round. The small rings of quad quandary allowed for numerous ways to manipulate them. I saw claws, "flippers," roller systems in all orientations, the traditional dual tank tread lift, the "stackerbot," 1114's variant on the stackerbot, and many, many other systems. Many of these robots were masters of their art, manipulating these rings with great dexterity. It may have looked "easy," but I know from having built a robot for that year that it must have taken months of nonstop effort. And, the end result really impressed the audience, because the end result was a quality product.

    On the other side of the spectrum is the atlas ball. For those who weren't around back then, Hanging-A-Round had this gigantic, 30-something inch diameter ball on the field, which gave a bonus for having it on your side. Most teams just passively rolled it around, or ignored it. But there would be the odd team that went above and beyond, and made a crazy expanding system to trap this ball in a corner, pick it up, or even toss it out of the arena to negate it's bonus. They didn't do the job too cleanly. But it was clear to the untrained eye that it took a great feat of engineering to even attempt this incredible task. The same applies to green eggs lifting up the mobile goals in Round Up, it was unquestionably impressive.

    The past two games have been somewhere in the middle. The objects have been large enough to make it difficult, and to limit the number of ways to pick them up. They make the designers think at first. Eventually, some people get decent at it, and people follow, because it's challenging enough to limit the number of effective ways to do this (see "NZ designs" from gateway). But, to the untrained eye, the task still isn't spectacular enough to inspire the awe that the Atlas ball did. No one walking into an event thinks that picking up the barrels is the hardest thing in the world. And yet, I still haven't seen a single robot with the ability to pick up a barrel in an arbitrary orientation, shift it to a desirable one, and score it reliably. The "best" robots juggle these things around in their intakes. It is certainty harder to play gateway competently than it is quad quandary. But the average quad quandary robot looks much more effective and reliable, and ultimately more impressive. 
  • Something that makes programming in autonomy during operator control highly desirable. 
  • I doubt it will be evident when we see the game this evening. But I want the opportunity for a game breaker. Because who doesn't love to go "Why didn't I think of that?"
  • A college challenge that demands two robots designed to work together. While I've made it clear I'm no big fan of gateway, I LOVE the twist the VRC GDC put on it for the college challenge. Here's hoping this trend continues.
  • Bring back the optional separate robot for the programming skills challenge. 
  • Minibots! Ok, maybe not...

And I'll close with something radical, that I think must be implemented in order to maintain VRC's credibility as a legitimate engineering challenge. 

A radical update to the game rules 6 months or further into the season, that requires teams to rethink things. 

I loved the year-long game when it was released. It has its advantages, particularly from an educational perspective, as teachers can use the game in their classrooms whenever it works for them. But from a competitive perspective, I'm beginning to wish VEX would go back to a defined "build season." Because a year has proven to be more than long enough to "figure out" a game. 

There have been a few ground-breaking innovations revealed on a large scale at world's every year. But those have been getting fewer and fewer each year. The first "NZ designs" hit VexForum before the summer was out, and the first six bar lift only took a few weeks. Elite were maxing out the Gateway field mere months after the curtain was raised. Even wallbots were in the public domain several months before, and the "super-stacker" was well-known going into worlds. In a competition that's about continuous improvement, with a platform that's versatile enough to allow low-risk complete rebuilds, eventually a point of diminishing returns is reached. Eventually, teams see something that they think is the best that can be done, and follow. The designs become stale, and winning becomes less and less about innovation. World's this year is filled with the same six bar+side roller design that was winning events 9 months ago. 

I want the teams that "figure it out" to have to prove themselves, and adapt to a changing game. A changing game is where matches are won and lost now...short of winning through superior design, the best teams now win through superior on-the-spot strategic decisions. I want to see VEX match this in their challenges, and match the truth of real world engineering, that your requirements are always changing. I want teams to be forced to adapt and improve, not only to better opponents, but to a living, growing game. 

This would be a difficult thing for VEX to do. FRC inadvertently did this to a couple teams this year, by closing a loophole regarding the definition of the bridge. The most well known team to have this happen to them was 3928. There was a lot of outrage towards this move by FRC at the time. How dare they change the challenge teams have worked so hard to solve?

But what did 3928 do? They adapted. They did a drastic, risky redesign of many aspects of their design, and came out better for it. They had a formidable robot at the midwest regional that was able to hold its own against some of the best teams in FIRST. 

This system wouldn't work as well in FRC, since it is very difficult to do anything other than "go with what you've got" in a six week build period. But to bring some added challenge into VEX, and force world championship caliber teams to work to improve, all year long. 

Monday, April 16, 2012

Something Awesome

How should you start your blog? With something awesome!


As the 2012 VRC championship approaches, I maintain that this machine from four years back is the greatest VEX robot ever created. Amazing concept and execution, topped off with an incredibly gutsy mid-season rebuild. This is the kind of thing I want to create someday.