Bert: You're making good points. I don't disagree. My quibble isn't with the idea of wheels vs legs, really, but with a top-heavy form factor that just happens to look like a bipod human/ape. If we wanted a strong, agile animal form, why not a cheetah or bear? But more to the point, why not a lower center of gravity? Those are the things that make me think the inventors were creating a marketing machine more than the most efficient robot.
Well, but the roads had to be invented too. So how useful is a robot, which might potentially be used in rescue operations, if it's optimized for smooth roads?
Wheels evolved from logs, used to move heavy objects along relatively smooth surfaces. But mules had to be used to move similarly heavy objects on difficult terrain. That's my point. Smooth roads didn't come with this planet, we had to invent them.
In fact, man makes his creations in his own image. Cars have two headlights, a nose, and these four wheels, the only practical alternative to arms and legs. I see this robot as an evolution of our creation. :)
I dunno. I think the wheel was invented as an improvement over lugging stuff around on two legs. I'm not saying robots shouldn't have arms. I just think a lower center of gravity and a better traction system would be more effective than a machine that is built in the image of his creator (how vain was that?)
" ... but a car with arm could probably climb it even better and move more efficiently on land."
Could it? A car with just one arm would have a heck of a time climbing a tree, with all that useless weight of wheels dragging it down. Honestly, think about it. If wheels were better or more flexible, for motion on this planet, why wouldn't nature have evolved that way in any species?
So, take this "car" and give it the four appendages all of the larger animals seem to have, instead of wheels, and what's the difference with this robot?
Even if many can't see past the "gimmick" aspect, it's hard to deny that making a machine with this form factor, one that operates well, is going to make a very flexible machine indeed. Way better than a wheeled vehicle, unless all you care about is top speed on smooth terrain. In the video, they did show it avoiding obstacles and adjusting for variable terrain underfoot, so it's not like they don't understand what's involved here.
Let's face it: robots are fun. They've fascinated all of us going back to Jules Verne. They inspire and they amuse. Carolyn asked what is DARPA marketing here? I think it is marketing DARPA with a message of: "Look, we're a fun bunch! Join us!" That message would be harder to sell (to most young engineers) with a lethal image like a smart bomb.
The Pentagon's mechanical man here doesn't make much sense, except as a marketing ploy. Do you agree/disagree?
That's exactly it--it is a marketing ploy. But, who are they marketing to? You an market the concept of fun, function, or even fear. If the functional aspect of this robot isn't high given its form factor for the type of jobs it is said to be made for, what is being marketed, and again to whom? Me, I still think that there is a frightening aspect to this robot. Why? As Tom points out, making it look like an agile atlete would make sense--and not cause further trauma to those humans it may someday be trying to save.
This is impressive although I would really like to see arms and hands that can manipulate small objects with touch feedback - say to fix a watch, or replace a screen on an iphone.
I agree the humanoid form is mostly for show as a snake would be much nicer for many jobs, although as has been rehashed many times over in movies and Science Fiction books - people like to look at people and would probably be creeped out dealing with other life forms although I am sure we would get used to it.
Power is important, but I think they are far away from needing a large energy pack although there are many other industries that can use one. I still have not seen any promising technologies with power density that would be practical for a robot, or much better than our current lithium products. Even nuclear would need to vent tremedous amounts of heat - I guess the robot cold have a large hat as a cooling tower... ;)
Bert: You're right that this hunk of metal could climb trees better than my car, but a car with arm could probably climb it even better and move more efficiently on land. Let's face it, they made it look like something from Lost Planet as a marketing gimmick. It just doesn't make a lot of sense compared to other forms. If they had to go with human, couldn't they have made it more like an agile athlete?
The quadroped called "big dog" made by boston dynamics has a gasoline fueled power house on board. I don't think power will really be the issue. Even other bipedal and fairly swift machines like the latest version of Asimo are able to carry a power pack and have been for several years.
The biggest hurdles are complexity of real time adjustment to environmental variables. When a piece of rubble rolls from under its foot, how does it react? If it encounters a drastic change in terrain, and ultimately a failure, can it recover? These are the problems they are really trying to solve.
Drones are, in essence, flying autonomous vehicles. Pros and cons surrounding drones today might well foreshadow the debate over the development of self-driving cars. In the context of a strongly regulated aviation industry, "self-flying" drones pose a fresh challenge. How safe is it to fly drones in different environments? Should drones be required for visual line of sight – as are piloted airplanes? Join EE Times' Junko Yoshida as she moderates a panel of drone experts.