Flying Robot Bird Unveiled

mikejuk writes "Festo, well known for their biologically inspired robots, have a new creation called SmartBird. It is amazing to watch and all the more amazing when you realize that it flaps its wings and all of the control is via a torsion drive which twists the wings during each flap. The whole thing depends on the constant intervention of the software to keep it under control."

Comments

Wow, just wow. The uncanny valley extends ....

[GNUALMAFUERTE]

Well beyond humanoids.

It's so beautiful, interesting, and yet creepy in a way.

Humanity's advances in certain areas (like robotics) are amazing. The sad part is that we are way ahead in certain areas, but way behind in other three key areas:

a) Energy:

We still have to crack the energy issue. We lack both reliable ways to gather energy, and reliable ways to store it.

b) AI

We are still in diapers in weak AI, and not even started in strong AI.

c) Economy

We are still based on the stupid principle of scarcity. Until we realize that we can produce as much as we need of just about anything, and that we are limiting ourselves by creating artificial scarcity to keep alive a system that's been dead for a long time, we won't make that breakthrough into what we thought the year 2000 was going to be.

Re:Wow, just wow. The uncanny valley extends ....

[physicsphairy]

Humanity's advances in certain areas (like robotics) are amazing. The sad part is that we are way ahead in certain areas, but way behind in other three key areas:

This is true, or not--it really is a matter of personal perspective. I myself think we are making fairly equitable process in terms of the limited resources we are splitting between all of our priorities:

We are still in diapers in weak AI, and not even started in strong AI.

Might as well be saying this in the year 1850 in respect to clockwork men. The human brain is not software so much as it is hardware, and modern transistor chips do not resemble it very well. I think our present level of AI is fairly suited to the sort of computers we are manufacturing. Get quantum computers (in which we are make quite excellent progress) up to where large chips are viable, and I think the code development of AI will be a few orders of magnitude more viable as well.

We still have to crack the energy issue. We lack both reliable ways to gather energy, and reliable ways to store it.

We can generate massive (nuclear) and store massive (dam) amounts of energy. Scaling it down becomes more troublesome. But the real issue is that it's easier to dream up ways to use energy than to produce it, due to troublesome and altogether uncircumventable laws of thermodynamics, which in one sense will always make us seem like we're lagging. Energy "does stuff," and we'll almost always want to "do as much stuff" as we can, which can always be translated into some kind of gain. Getting back to the AI issue: there are fundamental constraints on the the amount of computation that can be done per joule of energy expended. And will we ever be happy with the amount of computing we do? I don't think so. So, in that respect alone, we already have some desire for infinite energy, to say nothing of whatever finite supply we have at a given time.

We are still based on the stupid principle of scarcity. Until we realize that we can produce as much as we need of just about anything, and that we are limiting ourselves by creating artificial scarcity to keep alive a system that's been dead for a long time, we won't make that breakthrough into what we thought the year 2000 was going to be.

Humanity always produces as much as it needs. If not, people die and the equilibrium is re-established. It comes to producing as much as we want, and that, as given in one example, is pretty much limitless. Scarcity of resources in that respect is inevitable and the distribution of those resources is fairly well addressed by capitalism. I doubt there will be a better way to distribute them until our super-AI comes online to figure it all out for us. Until then, humans are provably terrible at guessing where resources should go by any means other than rational self-interest.

What? No Feathers?

[camperdave]

Why do no ornithopters employ feather analogs? They all have solid wings. Flight feathers form a check valve, letting air through the wing on the upstroke, and capturing it on the downstroke. This arrangement allows the bird to put a lot less energy into the upstroke (and thus not lose as much altitude). With a solid wing, you wind up pushing the craft down on the upstroke almost as much as you lift it on the downstroke.

Haven't you heard?

[degeneratemonkey]

About the word? Ba-ba-ba-... *ahem*... I, for one, welcome our new cybornithological overlords.

Re:Constant intervention of the software ...

[Kilrah_il]

And also, real bird also depend on constant intervention of the software, i.e brain. Usually when birds lose their heads, they stop flying. Or so I've heard.