3 ms·

The 30-Hour Shift That Turned a San Jose Robot Lab into a Global Spectacle

6 points by odysseyk 5mo ago

pocksuppet 5mo ago
Humanoid robots are publicity stunts. They are not well-designed for any task. Humans relate to them because they look like humans. That's it. Robot-shaped robots have been doing factory work for much longer than 30 hours at a time for decades. You don't need a humanoid robot to use computer vision to sort packages. Robots at mail sorting facilities have been doing that for decades too. The most effective way to sort a package is some sort of diverter that swings out diagonally across a conveyor belt, so the package falls off the side at that location, into a collection basket or onto another conveyor belt. Also, this article is paid advertising.
- skywal_l 5mo ago
  Yes but they are also a quick win to replace humans because they don't need accommodation or specific R&D. Any job by a human that has no been automated yet is because the ROI was too low to develop a specific robot and adapt the work environment for it. The human is basically the standard API. A humanoid robot is a drop-in replacement implementation of this API.
- solnyshok 5mo ago
  I have small scale 300sqm warehouse - incoming full pallets, outgoing mixed pallets. It is too small for industrial automation, but humanoid robot could be a good fit - unravel pallets, pick boxes, put boxes onto outgoing pallets and shrink wrap those.
Haven880 5mo ago
Nothing burger. Factories in Shenzhen fully automated with specialized robots running for the past 5 years. Generic humanoid robots not going to outperform those specialized robots.
- kgeist 5mo ago
  Generic robots are potentially faster to configure and set up in many different environments, and they're more accessible, no? If I wanted a specialized robot right now for my small business, I'm not even sure where to begin. But if I could just order a bunch of generic robots off Amazon and tell them what to do, that would be great. Sure, specialized robots would outperform them, but for many tasks, that would be overkill.
odysseyk 5mo ago
First of all, I want to say that this blog does not feature paid articles; it operates under a different paradigm. Of course, highly specialized automated machines, manipulators, and so on, designed to perform repetitive tasks, will generally win in terms of price, followed by speed and quality. However, this demonstration is intended to demonstrate the capabilities of humanoid robots, which will have a much wider range of applications, limited only by your imagination and the law. In 5-10 years, this will become commonplace; that's precisely what I'm talking about.
- Scipio_Afri 5mo ago
  "it operates under a different paradigm" And what paradigm is that which you decided to leave out ?
  - odysseyk 5mo ago
    [flagged]

6 points by odysseyk 5mo ago

Humanoid robots are publicity stunts. They are not well-designed for any task. Humans relate to them because they look like humans. That's it. Robot-shaped robots have been doing factory work for much longer than 30 hours at a time for decades. You don't need a humanoid robot to use computer vision to sort packages. Robots at mail sorting facilities have been doing that for decades too. The most effective way to sort a package is some sort of diverter that swings out diagonally across a conveyor belt, so the package falls off the side at that location, into a collection basket or onto another conveyor belt. Also, this article is paid advertising.

skywal_l 5mo ago
Yes but they are also a quick win to replace humans because they don't need accommodation or specific R&D. Any job by a human that has no been automated yet is because the ROI was too low to develop a specific robot and adapt the work environment for it. The human is basically the standard API. A humanoid robot is a drop-in replacement implementation of this API.
solnyshok 5mo ago
I have small scale 300sqm warehouse - incoming full pallets, outgoing mixed pallets. It is too small for industrial automation, but humanoid robot could be a good fit - unravel pallets, pick boxes, put boxes onto outgoing pallets and shrink wrap those.

Haven880 5mo ago

Nothing burger. Factories in Shenzhen fully automated with specialized robots running for the past 5 years. Generic humanoid robots not going to outperform those specialized robots.

kgeist 5mo ago
Generic robots are potentially faster to configure and set up in many different environments, and they're more accessible, no? If I wanted a specialized robot right now for my small business, I'm not even sure where to begin. But if I could just order a bunch of generic robots off Amazon and tell them what to do, that would be great. Sure, specialized robots would outperform them, but for many tasks, that would be overkill.

odysseyk 5mo ago

First of all, I want to say that this blog does not feature paid articles; it operates under a different paradigm. Of course, highly specialized automated machines, manipulators, and so on, designed to perform repetitive tasks, will generally win in terms of price, followed by speed and quality. However, this demonstration is intended to demonstrate the capabilities of humanoid robots, which will have a much wider range of applications, limited only by your imagination and the law. In 5-10 years, this will become commonplace; that's precisely what I'm talking about.

Scipio_Afri 5mo ago
"it operates under a different paradigm" And what paradigm is that which you decided to leave out ?
- odysseyk 5mo ago
  [flagged]