I’m pretty interested in autonomous vehicles, and because of the wide availability of cheap electronics and compute power, experimentation in this realm is increasingly within the grasp of amateurs. I like the idea of building small aerial drones which can not only follow a predetermined coarse, but can also work to avoid obstacles and the like.
Today I found a press release for a paper “Low-Power Parallel Algorithms for Single Image based Obstacle Avoidance in Aerial Robots”, which seemed right up my alley. A quick Google revealed this web page, containing links to their papers. The math involved is a little beyond the casual, but I think I’ll be able to sort it out. Very nifty stuff, bookmarked for later.
Oo, cool! As soon as I saw the video, I thought “wow, that sure looks like Cornell… is that Goldwin Smith Hall?!” And it is!
They sure didn’t have autonomous flying drones zooming around the Arts Quad when I was there, though. 🙂