I didn't realize until just now that I never posted a link to diary entries 3 and 4 on the main work page. Sorry, Jim. The links are there now, but hopefully you saw them before today...
Anyway, today is April 10 and there's not too much to add to the progress made in the last two weeks. The last few pieces needed to finish the assembly should arrive any day. Since I will need to make the AC adapter by soldering a couple wires together, I'm going to test out my soldering skills tonight by attempting to make a small cube out of resistors. If for reason I fail miserably at it, I think I'll just find someone else to help.
I'm not worried about completing the construction, but I'm a little nervous about the level of programming I'll be able to complete. At this point, though it looks as though the robot will at least be able to do some basic movements, but any higher level programming probably won't come until the last week of the semester.
While waiting for the last little pieces to get here I did a little background reading on imaging spectroscopy.
tetracorder One of of the newer systems used in satellite data image analysis is a sturdy set of algorithms they refer to as Tetracorder. Tetracorder takes huge arrays of spatially arranged spectral signatures collected by fly-by satellites; the data is analyzed for spectral identification by a decision tree that uses a set of multiple algorithms to determine the unique material composition of image. This method has been used to determine geographical attributes such as mineral composition, land structure, vegetation distribution (and even plant types and relative plant-water content), air moisture, and temperature.
The coolest thing I found in this paper about the Tetracorder was the documentation of the spectral diagnostic features that were used and found to be efficient. It's about time to meet with you, so I'll continue this summary after we talk.
