Gape in Awe at these Super Fast Robot Hands | Hacker News Reader

Gape in Awe at these Super Fast Robot Hands

singularityhub.com

19 pointskkleiner9 comments

Linked from the article is another on robot chefs: http://singularityhub.com/2009/08/03/the-robots-are-the-chef...

One thing that strikes me about this nascent technology is how much effort is being spent making robotic use of human hand tools.

These hand tools were developed in response to human morphology (handles on the ladels) and physiology (avoidance of contact with high temperatures). A robotic arm can be easily sterilized and has not need to avoid contact with boiling water. Hence, it seems ridiculous to have a robot use them instead of hand tools designed to better fit the robotic arm.

It only seems ridiculous if you assume the end goal was to have a robot be a chef or a baseball player. It was not. Those tasks are applications designed to answer the question: can we build robotic appendages that have the precision and dexterity of human hands?

In research, you see if you can get the simplest thing working first. Then you start adding complexity. In this case, the simplest thing is have a robot that does things human hands normally do.

Within the realm of the baseball playing robotics, this is true. There is a lot of research to be done on human locomotion and automated rapid response. Don't get me wrong, I was very impressed by the reproduction of a two finger fastball and the response by the batter.

But as for the cooking, the robots had to be rotated to their extremes to operate the ladle. The end result there was preparation of food, and the use of the ladle was not the simplest thing to have the robot do. That seemed out of place, and pointed at the ladle being the wrong tool for the job.

Hence, it seems ridiculous to have a robot use them instead of hand tools designed to better fit the robotic arm.

Legacy! We have a world full of legacy hardware designed for human use. It is in fact much cheaper to comply with the legacy.

So you can afford this (hypothetical) super-intelligent and versatile robot, but not the "pasta strainer" attachment?

What are you going to do with only a pasta strainer? You're going to need a whole kitchen's worth of attachments and the cost could quickly become large, especially if the appliances and fixtures need to be modified. Afterward the kitchen will be less useful for humans and the robot will still be useless for anything but kitchen tasks.

I have a robot that does nothing but clean my floor. It was still well worth its price. "Do one thing and do it well."

We already have kitchen devices that do one thing well: dishwasher, mixer, breadmaker, toaster, cuisinart, etc. At some point it will become cheaper to make one robot that can do it all.

It is more interesting to see them use human things. We can use them, so we know what they are doing.

It's also a lot harder. We already have machines which make food and are optimized for it.

It's just not cool in the same way.

Reply on news.ycombinator.com