So here's the low-down on Autoliv...
They do airbags mostly. Some other restraining stuff for car safety, but mostly airbags. There's a few of us who're interning there. I drew the long straw. I'm in the dynamic lab module. For laymen, I'm working on a sled that gets pulled with some 77' bungee cords (we use 14 of them), and smashed into random objects - poles, deformable barriers and the like. We use two high-speed cameras to record the action, and about 12-15 accelerometers. I take the data and integrate it to something useful to figure out our braking system. Which consists of a huge metal battering-ram (called the "thorn") that smashes through a whole bunch of steel bars. The bars end up getting shaped like "U"s (or omegas- for the greek). The sled weighs a little shy of 4000 lbs when it's fully loaded with the dummy.
Here's the thing I though was funny - we had to have another dummy shipped from Detroit. It came in yesterday, and it was all sorts of packaged up with bubble-wrap, plastic coating, protective foam... I stood there and thought "Were they afraid the truck hauling this thing was going to wreck? It's a crash-test-dummy for crying out loud!" Anyway, those things can cost anywhere from 80-200 grand depending on the instrumentation - so I guess it's a good thing to protect, eh?
So far, they've been having me work on the bungee acceleration (so we can hit target speeds of 20mph, 25mph, 35mph depending on the test) and the brake bars. The bungee acceleration should be a simple force balance F = m*a = k*x. The nonsense they teach us in school - the "k" is called the "spring constant". Problem is, from our data, it's not really a constant. No biggie, we're just going to have to make a chart and correlate it, right? Cinch. The brake bars on the other hand, would mess with the minds of many a professor. They're not too bad when they're supported with only two points and the thorn hitting the center, but they're not. They're wedged between four points - which takes the "simply supported beam" theory (the only stuff they teach in school) and screws it all up with some crazy physics. So I decided I'd forget theory and just use our data to get some correlation. I'm currently hunting down the relationship between bar thickness (when stacked or staged) and average bar energy absorption rate. Sounds fancy, but all it means is that the bigger bars bend less and slow down the same.
So there's a tid-bit of the nonsense I do. I've also made a few boomerangs in strange shapes. One is shaped like the Aggie "A". It works. I'm that good.
I've been running. Slow and short.
Haven't married yet.