Lots to be said here!
First, I'll say thanks to HAD, for the Teensy and LightBlue Bean! I'm not sure what to do with the Teensy yet, but I've got a few ideas for the bean. (Including a "does this work the way I think it does?" project.)
First off, a huge thanks to Grady Houger for answering some questions about the farming side of things and helping me get a good idea of priorities in terms of what sequence to worry about things in. He pointed out a few organizations I should contact. (All of which I still need to contact. I'm crazy skittish about talking on the phone.)
I've decided to break the project into several phases, which I expect to be able to move on quickly, but each stage is meant to be sort of self contained:
- Phase 1 is simple proof of concept. Since I can't find enough meaningful data on the LIDAR based scans, and the same information seems to be obtainable through Multispectrum Photography, I've shifted to using a multispectrum webcam. (On which, more data is available.) Ideally phase 1 will end with a flyable prototype that can measure ground moisture and provide NDVI info at the same time by WiFi to a base station.
- Phase 2 will explore maneuverability, and adding propellers to the LTA system to provide greater coverage.
- Phase 3 will explore things of greater magnitude that will probably not be finished by August, such as weed identification, and possible delivery of targeted fertilizers/pesticides/etc.
Phase 3 is my greatest concern, as plant identification is going to be challenging.
I've been working on sourcing parts. I've already mentioned wanting to use a Helikite for prototyping. Seeing as how I need to know how much weight I will need to carry to order my helikite, I have elected to deal with that once I get my initial sensor setup prototyped. I've found a reasonably priced Infragram webcam available from Public Lab, along with a host of NVIR photography information pages to dig through. With any luck the webcam and spare filters will be here monday! I plan on hooking the webcam up to a RasPi, at least for the short term.
Still needing to be sourced are Solar cells, good battery cells, a good low power GPS receiver, and possibly a few other things. I want to mount the webcam in one of those gimbled housings that can be panned and tilted around to look at more than just forward/down, so I will need to build or purchase the housing and work out motors for it. (The bean's accelerometer may help, here.)
This gets me to the latest HAD prize news posting, about the Shapeways gift cards.
As I need to kludge together an equipment housing, and I have been wanting a 3D printer for a long time now, I decided this was the time to purchase one. It arrived a week ago and the first two test prints I've done have come out perfect. If HAD awards me the card, it will go to printing the "final" housing, as I think the SLS prints from Shapeways will look better and more polished than a print from my own dual extruder unit.
With that said: If it comes down to awarding the gift card to my project, or to another project that doesn't have a 3D printer available, please, give it to the other guy. :)