3 ms·

> A better automatic appeltizer will change the world in a more meaningful way than a humanoid robot. Is this a joke? An actually functional humanoid robot wou

> A better automatic appeltizer will change the world in a more meaningful way than a humanoid robot. Is this a joke? An actually functional humanoid robot would be able to be dropped into basically any factory in the world and do a job that most humans agree sucks. And you think that is less impactful than building an expensive and very specific tool for a specific job that humans are already not doing very much?

dragontamer 4y ago
> An actually functional humanoid robot would be able to be dropped into basically any factory in the world and do a job that most humans agree sucks Oh really? Can this Tesla bot pick up a strawberry without crushing it? Can this Tesla bot identify a weed in a yard, and uproot the weed without harming the other plants? ------ Weeding and harvesting of fruits are just two, labor intensive human tasks today requiring the human touch. There are also other human tasks: opening a door, walking up and down stairs, folding laundry. Etc. Etc. What can this TeslaBot do? As I noted earlier: Boston Dynamic's 'Spot' is picking up landmines in Ukraine. -------- But we know how agricultural robots look like. https://youtu.be/8BAQCpUu3ww https://youtu.be/8BAQCpUu3ww Is humanoid even an advantage? No. It really isn't. Humanoid unnecessarily hinders your imagination to the possible efficiencies at play here. Honestly, a weeding robot might just be a powerful laser + quadcopter drone. No reason for it to be human at all (bipedal, fingers, etc. Etc). Or maybe some chemical herbicide/poison applicator (spraying Hydrogen Peroxide or something like that?)
- fastball 4y ago
  What the current TeslaBot can or cannot do is not the point I was making, I was merely responding to you assertion about the transformative potential of humanoid robots. Humanoid is an advantage because all of our existing infrastructure is geared towards humans – hands with five fingers / two legs / etc. Again, that means a high-functioning humanoid robot could start doing shitty manual labor people don't want to do overnight. This is obvious in a factory because you don't need to retrofit your entire factory, but even more obvious in a context like a home, where there are loads of small activities it would be nice to automate (gardening/washing/cleaning/etc) that would each require its own robot if you don't have a humanoid one, at which point... doesn't it make more sense to just have the one humanoid robot?
  - dragontamer 4y ago
    > What the current TeslaBot can or cannot do is not the point I was making Yes it is. > Humanoid is an advantage because all of our existing infrastructure is geared towards humans Cool. So can Teslabot open a door? If not, when do you think Teslabot can open a door designed for humans to open? > Again, that means a high-functioning humanoid robot could start doing shitty manual labor people don't want to do overnight. Only after that humanoid robot can open the door and climb the stairs and reach the factory floor. Alternatively, you *design the factory around the robot* instead. You know, so you can do things like... https://www.youtube.com/watch?v=S8qkaTsr2_o https://www.youtube.com/watch?v=S8qkaTsr2_o ------------ Modern infrastructure, in the USA at least, is designed for machines, not humans. We have pallets to move objects around, forklifts to move those pallets, bays that directly lead onto trucks. You don't design around humans. Humans can only lift maybe 50 Lbs of weight (25kg) safely/reliably without injuring themselves. A good pallet-jack can lift 5000lbs and be moved around the factory no problem. Interfacing with the "machine" system of pallets and pallet jacks more quickly/efficiently is the meta. That's why an automatic palletizer / unpalletizer is a bigger deal than any humanoid robot Tesla can make. https://www.youtube.com/watch?v=sXkEjbZtNic https://www.youtube.com/watch?v=sXkEjbZtNic

dragontamer 4y ago

> An actually functional humanoid robot would be able to be dropped into basically any factory in the world and do a job that most humans agree sucks Oh really? Can this Tesla bot pick up a strawberry without crushing it? Can this Tesla bot identify a weed in a yard, and uproot the weed without harming the other plants? ------ Weeding and harvesting of fruits are just two, labor intensive human tasks today requiring the human touch. There are also other human tasks: opening a door, walking up and down stairs, folding laundry. Etc. Etc. What can this TeslaBot do? As I noted earlier: Boston Dynamic's 'Spot' is picking up landmines in Ukraine. -------- But we know how agricultural robots look like. https://youtu.be/8BAQCpUu3ww https://youtu.be/8BAQCpUu3ww Is humanoid even an advantage? No. It really isn't. Humanoid unnecessarily hinders your imagination to the possible efficiencies at play here. Honestly, a weeding robot might just be a powerful laser + quadcopter drone. No reason for it to be human at all (bipedal, fingers, etc. Etc). Or maybe some chemical herbicide/poison applicator (spraying Hydrogen Peroxide or something like that?)

fastball 4y ago
What the current TeslaBot can or cannot do is not the point I was making, I was merely responding to you assertion about the transformative potential of humanoid robots. Humanoid is an advantage because all of our existing infrastructure is geared towards humans – hands with five fingers / two legs / etc. Again, that means a high-functioning humanoid robot could start doing shitty manual labor people don't want to do overnight. This is obvious in a factory because you don't need to retrofit your entire factory, but even more obvious in a context like a home, where there are loads of small activities it would be nice to automate (gardening/washing/cleaning/etc) that would each require its own robot if you don't have a humanoid one, at which point... doesn't it make more sense to just have the one humanoid robot?
- dragontamer 4y ago
  > What the current TeslaBot can or cannot do is not the point I was making Yes it is. > Humanoid is an advantage because all of our existing infrastructure is geared towards humans Cool. So can Teslabot open a door? If not, when do you think Teslabot can open a door designed for humans to open? > Again, that means a high-functioning humanoid robot could start doing shitty manual labor people don't want to do overnight. Only after that humanoid robot can open the door and climb the stairs and reach the factory floor. Alternatively, you *design the factory around the robot* instead. You know, so you can do things like... https://www.youtube.com/watch?v=S8qkaTsr2_o https://www.youtube.com/watch?v=S8qkaTsr2_o ------------ Modern infrastructure, in the USA at least, is designed for machines, not humans. We have pallets to move objects around, forklifts to move those pallets, bays that directly lead onto trucks. You don't design around humans. Humans can only lift maybe 50 Lbs of weight (25kg) safely/reliably without injuring themselves. A good pallet-jack can lift 5000lbs and be moved around the factory no problem. Interfacing with the "machine" system of pallets and pallet jacks more quickly/efficiently is the meta. That's why an automatic palletizer / unpalletizer is a bigger deal than any humanoid robot Tesla can make. https://www.youtube.com/watch?v=sXkEjbZtNic https://www.youtube.com/watch?v=sXkEjbZtNic