Well, it appears we’ve stalled for the night on our crater-avoiding algorithms.
Chris was working on them all day, with Eric jumping in at the end of the night to fix a rather unpleasant bug that had the rover actively seeking out craters and martians. Unfortunately, Chris had to leave at ten to meet someone, Eric had to leave for the train shortly thereafter, and I didn’t realize this until about 9 p.m., so I didn’t have time to get up to speed on what they were doing before they were gone.
It was an hour or two after they’d left that I (and later Bryan, who joined me after he’d finished up creating a bunch of new and interesting maps for testing) started to get enough of a handle on the code that we could be mystified by parts of it. It may well all be good code, and it might be just that we’re unknowledgable and confused, but there’s a definite possibility that it’s the product of tired minds. The fact that it still runs into boulders and craters is, frankly, not a good sign.
And the topper? Neither Chris nor Eric are in tomorrow until the evening, after work. (The last day of the contest is on Monday, in Tokyo, with the submission deadline at 4 a.m. Tuesday morning.)
Well, Paul is coming (for the first time) tomorrow, and he says he knows a good bit about computational geometry, so I’m hoping he’ll come up with some fantastically ingeneous algorithms that will solve all of our problems.
In the meantime, I’ve parameterized our destination point generator, and we can at least submit our previous “Full speed ahead, and damn the craters” strategy.
Sigh