It has been a long week, and I’m glad it’s Friday. Today turned out well. We had a meeting this afternoon with some EX and EG guys to talk about our ideas for using neural networks as an abort determiner (or determinator as some call it, but Grammar Girl maintains that determiner is the correct word) for the 2009 Mars mission. It is a cool idea, and I presented our charts on it. I’m not very good at giving the presentation yet, because I am still learning about neural nets and how we plan to use them myself, but fortunately I had a lot of backup at the meeting. And it was less of a presentation than a roundtable discussion.
Kara, our co-op from last fall, proved that neural networks could work as an abort determiner, but she was only studying a very specific case–trying to land during a severe dust storm on Mars. So this summer it is my job to broaden the scope of the project. Anyway, the end result of today’s meeting is that they want us to continue developing the neural network technology and revisit the problem in a few months to see how it’s doing with a broader range of Mars entry trajectories.
So. I know what I will be doing at work for a while now, and that’s a good thing. As I learn more about neural networks, I realize that they are really pretty interesting, and so even though it’s more code, maybe it won’t be so bad. And, I get to write the code, rather than just debug endlessly.
I’m pooped. Time to go home.
