Autonomous end game

[dtengineering]

I say lets give the most dramatic moments of the competition over to the programmers. I am confident that the GDC can develop a game that addresses much of the concern listed here... however there are ways of adding to the importance of sensors and programming other than just changing the auto mode.

For instance consider a game where you are trying to put objects into four different goals. At various times in the game an IR beacon would turn on over top of one of the goals, thus doubling the value of any balls deposited during that period. Drivers would not be able to tell which goal was being doubled and when unless the robot told them. Or perhaps the goals only open at certain times and the IR beacon goes on 10-15 seconds ahead of the goal opening so that robots that sense IR can get lined up to score... or defend... first.

Or perhaps a game that involves a great number of very small game pieces.... coloured golf balls for instance... that need to be sorted in order to receive maximum bonus points. There would be so many pieces that there would be no way the drivers could sort them fast enough, so it would have to be done autonomously on-board the robot in order to have a dominant machine.

If there is sufficient incentive to develop a good auto mode, then teams will allocate their resources to making it happen. I know this year had we decided to focus on auto we could have easily dropped mecanum drive for a repeat of our quite successful 6wd platform from last year and gained at least a week of build time for programming and testing auto, rather than drive code.

The last big change to the game format was going from two to three robots. I think it is time for another change... or changes (how about a slightly larger playing field) to shake things up a bit.

Jason