This is not technically a computer game-programming question, but I thought maybe a physics person could help me out. I'm building and programming a robot to navigate an obstacle course for a mechatronics class I'm taking this quarter (cool stuff, see me118.stanford.edu for final projects from other years). Some of the obstacles include small lakes fill with water. It occurred to me that I wouldn't need to bother trying to navigate around these lakes if I could use liquid nitrogen to freeze them so that my robot could simply roll over them. These lakes are about 2 feet long and 8 inches wide and 3 inches deep. I estimate it's about three liters of water. I was wondering if anyone could give me an idea of how much liquid nitrogen would be required in practice to freeze this amount of water so that a 5 pound robot could roll over it. I've been poking around online and all I've been able to find is that the specific heat of LN2 seems to be a quarter that of H2O, so I would need more than 12 liters of LN2 to freeze one lake (since there is also the latent heat of fusion to consider). This is a lot of LN2. I was wondering if anyone here on the physics forum has practical experience working with LN2 (I've only played with it myself) and if so, could they comment on whether this plan is likely to work. Other issues I've considered is the amount of LN2 my robot can reasonably carry, since the robot can't be larger than one foot in all dimensions and most vacuum flasks I've seen are larger than this. I'd also be worried about the condensation messing with my electronics. Is there a better freezing agent that I could use? I settled on LN2 because it doesn't explode like liquid hydrogen might and it's much colder than dry ice.

Instead of carrying 12 liters of LN2, why not carry a 3 liter jug and vaccum out the water. perhaps you could name your robot 'moses' then :)

Quote:Original post by Fragmo
Instead of carrying 12 liters of LN2, why not carry a 3 liter jug and vaccum out the water. perhaps you could name your robot 'moses' then :)

Why bother carrying a jug to suck it into? why not just suck it up and spray it out behind the robot?

You have to detect the water wether you go around or freeze it. Why not just go around?

what if the robot gets thirsty though?

Pumping it out would be a great solution except that the robot is not allowed to touch the bottom of the lake. We also have been toying with the idea of filling it with sand or something like that. I do like the thermite idea, though. How much thermite would be required to boil 3 liters of water? I know that stuff is crazy hot, but I bet it's a lot.

Simplier ideas like going *around* the lake is what most groups are doing. We want to do something a little bit more impressive since we are engineering hotshots who feel the need to show off.

Extendable bridge? Have it come out the top of the Robot, then drop it, drive off. Or, you could have it come out the bottom, and have it drive across on tracks, and then un-extend it(that way, it keeps the ladder). Just throwing out ideas.