3 ms·

>> Not to dismiss the challenges with automating burgers, but its pretty straightforward compared to self-driving tech - there's basically no unknowns! I alway

>> Not to dismiss the challenges with automating burgers, but its pretty straightforward compared to self-driving tech - there's basically no unknowns! I always assume things are more complicated than they appear on the surface, so this is a surprise. You sound like you've explored this topic thoroughly. What is involved in having a robot produce a hamburger (not just the patty, of course)?

pigscantfly 8y ago
OP is referring to the likelihood that a blizzard blows through the prep kitchen as opposed to the likelihood of a hard rainstorm on the 101 during the morning commute. Self-driving cars have to reliably cope with a wider range of uncertainties than a robot cooking burgers in a carefully controlled (by comparison) environment.
spunker540 8y ago
I’m not saying it’s easy- there’s probably a reason no one’s done it yet. But the comparison seems a little off. There’s no decades long darpa research grants required, and probably very few phds. There’s no predicting what other humans will do, there are no lives at stake (other than food safety precautions). The amount of capital alone that’s been invested in self driving tech (and it’s still not good enough) compared to the amount invested in burger tech makes pretty clear how hard one problem may be. This is basically a precision automated assembly line, and to be honest plenty of things just as complicated as burgers have been assembled automatically.
HeyLaughingBoy 8y ago
Process automation (a field I've been involved in for around 20 years) is vastly easier because the environment doesn't change by much. You can pretty much say "move this thing over here, verify that you moved it over there, add this chemical, and mix, verify that the chemical was added and then mixed" and be pretty confident that it will work in all likely scenarios. Self-driving cars, OTOH, have to work with nice smooth roads, or potholed monstrosities, or ice & snow, deer jumping out right in front of the car, or idiots tailgating you at 70 mph while you're trying to maintain distance from the car in front of you. Their environment is always changing, and not generally in a predictable fashion.

pigscantfly 8y ago

OP is referring to the likelihood that a blizzard blows through the prep kitchen as opposed to the likelihood of a hard rainstorm on the 101 during the morning commute. Self-driving cars have to reliably cope with a wider range of uncertainties than a robot cooking burgers in a carefully controlled (by comparison) environment.

spunker540 8y ago

I’m not saying it’s easy- there’s probably a reason no one’s done it yet. But the comparison seems a little off. There’s no decades long darpa research grants required, and probably very few phds. There’s no predicting what other humans will do, there are no lives at stake (other than food safety precautions). The amount of capital alone that’s been invested in self driving tech (and it’s still not good enough) compared to the amount invested in burger tech makes pretty clear how hard one problem may be. This is basically a precision automated assembly line, and to be honest plenty of things just as complicated as burgers have been assembled automatically.

HeyLaughingBoy 8y ago

Process automation (a field I've been involved in for around 20 years) is vastly easier because the environment doesn't change by much. You can pretty much say "move this thing over here, verify that you moved it over there, add this chemical, and mix, verify that the chemical was added and then mixed" and be pretty confident that it will work in all likely scenarios. Self-driving cars, OTOH, have to work with nice smooth roads, or potholed monstrosities, or ice & snow, deer jumping out right in front of the car, or idiots tailgating you at 70 mph while you're trying to maintain distance from the car in front of you. Their environment is always changing, and not generally in a predictable fashion.