Well, I haven't posted much in ages, so a quick summary...
Over the past 15 months, I've been working on pretty much the equivalent of a Computer Science degree, unaccredited and online... So, my (frame) building has taken a back step.
I'm still counting this as progress towards building though, as I'm doing these courses to help on the "autonomous" idea. Seriously, when some big names and big universities are offering to help you understand what to do, and how to do it, you sit down and listen.
So - find some time, get on Coursera, Udacity, and edX (plus whatever other good ones are around), and start learning things.
This is, of course, assuming that you don't just want a static model, or radio controlled... and I don't.
Not to say that I haven't been doing some hardware related stuff. My RaspberryPi is encased, I'm currently waiting on some analog-to-digital converter chips, and some input/output chips that interface very, very easily to the Pi. The RPi (if you weren't aware) has an input/output header on it, giving you some interfacing - the chips make it easier to do so. The only problem is that the RPi runs at 3.3V, rather than 5V so you have to allow for that.
I've also taken part in the crowd-funding of Adapteva's Parallela boards - 16 core processor, easy to program and interface. Hence, currently taking a course in parallel programming. Particle filtering can really benefit from parallel processing. And if the mathematics on the wiki page are indecipherable, you should really check out Sebastian Thrun's Robot Navigation course on Udacity.
What else... I have found about $3,000 worth of chips and sensors that I want. I've fallen in love with the listed capabilities of CMUcam v4 - so I've ordered one... and am thinking about a second, which would mean stereo vision, and making passive visual ranging possible.... I've also found a laser ranger that could be worth the cost and effort of getting... Of course, this means I have three small cameras... and am moving towards an R5 head... Just going to need plenty of computing power... So maybe another RPi or two...
