Though it can at times be overwhelming, one of my favorite things about working on a complex project is the vast number of things – some small, some large – that one has to learn to make it all work.  For this phase of the project, we procured a plant stress camera, which is destined for the ArduPlane and perhaps the quadcopters as well.
ImageJ
The camera – a Canon SX230 HS modified by LDP LLC/MaxMax.com – arrived recently, and I fired it up today to take some pictures of plants.  I’ve also been reading up on ImageJ, a public domain, Java-based image processing program developed by the National Institutes of Health here in these United States.  I don’t know much about it yet, but the software is extensible via plugins and macros, and the vendor provided various macros to convert RGB images into color scaled Enhanced Normalized Difference Vegetation  Index images, or ENDVI.

The camera also needs to be updated with the Canon Hack Development Kit (CHDK), custom firmware that enables, among other things timed, continuous shooting, which is critical for our purposes.  I’ll be updating the camera with that just as soon as I can read up on it.  Onward!

Here are the D2L usage figures for Fall 2012:

D2L Sites = 350
Graded Components (FLC course sections linked to D2L Sites above) = 426
Faculty Using D2L = 150
Student Enrollments = 12,643

The difference between D2L Sites and Graded Components indicates that some faculty maintain a single D2L site for multiple sections of the same class.  For instance, if a faculty member teaches 3 sections of Nutrition 300, they might choose to set up a single D2L site for all three sections.

Here are charts showing FLC’s D2L usage across the above four dimensions from Summer 2008 (when D2L first became available) to the present:

d2l_usage_01

d2l_usage_02

Friday’s flight ended in a tree, but in the seconds before that crash, I had the idea for the first time that I might just be able to fly this thing.  I thought about it all weekend, and resolved to never again fly without the GoPro.

This morning, I strapped the camera to the bottom of the aircraft, and headed down to the loading dock behind the bookstore.  I’m actually starting to get the hang of this thing, though I am still seized with a tiny bit of panic when first it jumps into the air, and sometimes when it seems to make its own decisions.  Nevertheless, I was able to execute seven or eight successful – successful here being defined as flights that don’t end in repairs – takeoffs and landings.  Here’s video from the best one:

To do:
Tweak the camera settings to limit the width of the angle
Come up with a better way to secure the camera to the quadcopter – as it is, I’ve zip-tied it to the bottom, but have to waste a tie each time I remove the camera.

Kevin!Spent the afternoon out at Parallax in Rocklin.  Kevin and Nick and Ken flew various iterations of the ELEV-8 platform, and thrilled the gathered crowd with feats of derring-do.  Balloons were popped, streamers were clipped, and one flying machine landed in the drink.  A good time was had by all.

After the demo, the FLC contingent – Dan Ross (CIS and Engineering), Dave Myers (student and president of the recently chartered FLC Robotics Club) and I toured the facility, and I had the chance to talk to Nick about some of the particulars of FLC’s project, and to say hello to Jessica of Dweeb Den fame.  Can’t wait to get the quadcopters in and get them flying!  Below is a video of a two copters dogfighting, and here’s a link to the complete set of photos from today’s event.  Thanks to Jim and and the whole Parallax crew.