Tree Climbing Robot

galactic grub writes "New Scientist's new Tech Blog has an article about a remarkable, if slightly creepy, tree-climbing robot being developed by robotics experts from Carnegie Mellon and several other US Universities. The article comes complete with a video clip of it going up several different surfaces."

Comments

Re:Still can't beat the japanese

[MaWeiTao]

Why exactly do you say that?

Sure Japanese have developed some impressive robots, but I wouldn't call something like the ballroom dancing robot a great feat of technology. Japanese designers seem to go for flashy robots, putting immense effort in creating something that has little practical utility but creates quite a stir. One company developed a humanoid robot and then we see dozens of companies cloning the original concept.

The ones developed in the US and Europe tend to be developed for real world applications. They don't look pretty, but they get the job done, solving a specific challenge in the process.

Not to discredit what the Japanese are doing, as they certainly are innovating too, but there's no reason to put down this work just because it doesn't look like Honda's ASIMO.

Follow up

[BecomingLumberg]

Any kid knows that getting up is easy, but getting down is much harder. How have they faired on that?

Re:Still can't beat the japanese

[PitaBred]

So, the AI lessons in making a robot react CORRECTLY to a ping-pong ball have absolutely no relationship to things like #3? They're solving 'problems' while gaining massive leaps in understanding. Even if their end result isn't immediately useful, the lessons they learn from it are.