4 ms·
I'm starting to think that any sort of general solution would have to be extremely complex... The optimum path would be something like a function of itself and
by fullstackchris 8y ago
I'm starting to think that any sort of general solution would have to be extremely complex...
The optimum path would be something like a function of itself and the shape of the forest. That is, there is probably a general solution to the INITIAL 'optimum' path based on the forest shape, but then your optimum path changes at any nth 'step', because as you haven't found the border of the forest yet, you learn a bit more (at least probabilistically) about where you could be in the forest...
I think that makes sense, but I have no idea how a solution method like that could be presented analytically.
- asah 8y agoIIUC you know the shape apriori. It's still hard.
- SilasX 8y agoIIRC in practice you can exploit a regularity that's common in the real world, which is that "people live near flowing water". So, just identify the downhill direction and follow it to a stream/river, which will guide you to people.
- caf 8y agoThis also works as a "lost in an ocean" problem as well, though.
- de_watcher 8y agoThe solution to each forest is a curve with the 'start' and the 'end' point. We minimize the length of the curve for that forest. Constraint: if the starting point of the curve is placed anywhere in that specific forest then some point of it must be outside regardless of the orientation of the curve. That's for avoiding thinking about the function of itself.