The big weekend is finally here. Jane and I are in Hartford for the weekend for the Trinity College robot competition. So the robot is officially "done". Of course by "done" I mean I am out of time! There are still a lot of things I would like to do, mostly on the software end but I am still fairly happy with what I have. The robot is able to get around the maze without getting stuck. It is able to put the candle out nearly all of the time, the only "glitch" here is sometimes it bumps into the candle before it sees it. I can find the "baby" somewhat reliably although that hasn't been tested quite as much.
The biggest thing I wasn't able to get to is some sort of mapping software so that it has some concept of where it is at. It is pretty simple navigation code at this point. It simply follows the left wall then after a certain amount of time (which can be configured at compile time) it switches to following the opposite wall. Hopefully this will be enough to prevent it getting stuck in a loop. Unlike previous years they are not going to reconfigure the maze between runs so I hopefully can look at the layout early in the day and make any needed changes.
I am told there are a total of 10 robots entered in the "expert" division. I talked with 3 or 4 of the other "expert" division builders today. It sounds like most of them made the typical mistake of making things too complex and leaving too much to do toward the end. My robot was the only one I saw actually moving about reliably in the maze so it is looking good as long as nothing breaks (knocking on wood as I type of course).
They usually post results on the web page later in the day after the competition if any of you are curious to see how I did. I will certainly add a new post here but likely not until Monday and since that is my last day before going back to work I will probably sleep late!