Slashdot Mirror
High School Robotics Competition Kicks Off
DeviceGuru writes "Some 35,000 high school students from over 1500 high schools in eight countries today began competing in the annual US FIRST student robotics contest. This year's competition, dubbed FIRST Overdrive, challenges the student teams to build semi-autonomous robots that will move 40-inch diameter inflatable balls around a playing field and score the most points. In this year's game, two alliances of three teams each work collaboratively to win each round. An animated simulation of the game (in several video formats) is available online."
Why enter Robocup?
I'm sure it's a pet fave of yours, but why might it be better for high school students who don't know much about engineering in the first place?
FIRST team do a new robot every year. Makes it easier for students to get in to. FIRST has the coordination for over a thousand teams to compete in roughly 40 regional competitions. FIRST robots are barely autonomous... generally the first 15 seconds of every game has been the autonomous mode. The rest is been teleoperated. FIRST provides teams with a starting kit for each year's robot, and standardizes key components, such as the motors. This fosters a community of sharing how-to info in using and troubleshooting most of the robot.
Ultimately, FIRST robots aren't very complex (though they can be for a team with a lot of support and organization). A bipedal soccer playing autonomous robot is a lot more complex of a project that I would have been willing to jump into in high school.
It will be interesting to see what type of schools in what countries do better. I participated in the Panasonic robotics competition held in NJ when I went to high school. A majority of the time, it was the private schools that end up doing the best.
The new hybrid period at the beginning of the match is where we'll really get to see teams "shine" (pun intended); You see, in the past teams usually just programed a direct control system and maybe some of the more savvy teams did some dead reckoning for an autonomous mode- Now teams are going to have to figure out just what predictive programming is, and are going to have to design their own method of conveying commands. Personally I'm expecting the majority of teams to use IR, but as I said, some teams are going to be very clever and use something completely unorthodox, and that's the best part of the competition imho.
When I was in high school in the 1950s, I participated in engineering contests like these, at least until they canceled them. Back in those days, they usually had us build a structure of some sort. Bridges were a common one, with the goal being to build a bridge that could hold a certain weight, but was made from only certain materials and the bridge itself had to be under a specific weight. In most cases, it was possible to win the competitions just by doing a good job gluing the balsa wood together, and just using the triangle shape.
But the kids nowadays must know complex fields like robots and software development. Even with simplified AI, it's still no doubt a real challenge. And to bring collaboration into the mix makes it even more difficult. Looking back, those of us building bridges had it really easy.
I still recall why the canceled our bridge building contests. One of the fellows at one of the other high schools in the district somehow managed to drop a 15 pound weight onto his scrotum while trying to attach it to his bridge. We all thought he was partaking in horseplay and got what he deserved, but apparently the contest organizers thought differently, and canceled all such contests.
Being a student at a high school which is participating for the first time this year, I attended a seminar on the rules, regulations, etc. Here's a summary of what seemed interesting:
The competition starts with an automated "challenge," in which the robot must do without human intervention. It then progresses into the manual portion of the competition, in which you can have whatever blend of automation and manual control you wish.
You're limited to the parts in the kit, plus an approved list of parts from third-party vendors. This is apparrently to avoid, for example, someone adding a ridiculously-powerful motor and accidentally running a 200-lb "robot" through the plexiglass around the areana and into the stands or operator booths.
The embedded computer is based on a PIC. Apart from that, it basically contains a wireless reciever of some sort to allow an official to shut down all robots in the areana in case Something Bad(tm) happens. You can add electronics if you want, but all connections to motors must be through the provided system.
Programming is usually done in C; Microchip's SDK is included in the kit. There's apparrently a small library containing functions for motor control and such things. You get a 25hz runloop, from which you can schedule your own functions to run.
Motors are connected through solid-state PWM drivers. The PIC provides PWM for you, but a PID controller (which is pretty much essential) must be implemented in software.
How to enable garbage collection on a system without protected memory: #define malloc() ((void *) rand())
Actually, most years the initial autonomous period (which was only like 10-15 second long...) was really completely independent, but this year we can use tv remotes and a provided IR receiver to have limited control of the robot even during this period. These really aren't "robots" anymore, but just complicated RC cars with arms. I'm not complaining though, my team has enough problems without having to worry about programming the thing. All we have to do is reconfigure the wheel-joystick relationship and we're set.
As a mentor I contribute my time and expertise for free to a local school who has competed for several years. I can say it is both highly stressful and highly fun at the same time. Both the high-school kids and mentors enjoy it a lot and it is really intense. I have seen several kids who had never previously considered messing with this "sciency stuff [sic]" get involved and completely change their minds about what they thought was cool and what they were going to consider going to college for. I am also very impressed by the solutions that are successfully built for the challenges. Dealing with the time limits, scavenging for parts, keeping things within specs, and conceiving/building a successful design would be a challenge even for adults. Its a great real world like experience for the kids.
I was the chief programmer on my team when I was in high school and every year I saw that programming became less and less important to the event and it was more engineering and marketing based. Its really disappointing and this year sounds like they just build a robot and drive it around, autonomous is becoming less and less important and its no longer 100% autonomous this year.