iRobot says its Warrior bot is more human than ASIMO
By Donald Melanson 
posted April 15th 2009 1:12PM
posted April 15th 2009 1:12PM
Is a humanoid robot inherently more human than your average task-specific bot? iRobot apparently doesn't think so, and it's taken square aim at one of the most popular humanoid bots around to make its point. That argument arose in a recent interview with CNET News, where iRobot CEO Colin Angle responded to a question about humanoid robots by flatly asking, "why would you want to make a humanoid robot?," adding that they might be good for movies or those looking for a robot companion, "but other than that, most tasks are best tackled by designs that are not constrained by trying to look like a person." He then goes on to note that Honda's ASIMO "requires a team of 10 or 15 people to maintain it, it can walk about, maybe, half a meter per second and in some situations climb stairs over the course of a few minutes, and if it ever falls down, it's a paperweight." On the other hand, iRobot's Warrior bot, he says, "can take a 10-foot drop onto concrete, drive 20 miles an hour, drive up stairs without stopping at full speed, carry 200 pounds of payload and has, maybe, five motors," which means it can "go nearly everywhere a human can." So, Angle says, "you look at these things and say, which one of these is a robot human? The answer is, Warrior." Is it? Or, deep down, is there a little bit of ASIMO in all of us?
garbage/poop
Is "Colin Angle" even a real name?
"can take a 10-foot drop onto concrete, drive 20 miles an hour, drive up stairs without stopping at full speed, carry 200 pounds of payload and has, maybe, five motors,"
wait, is he saying we can do this? Are people working at iRobot superhuman???
A robot with legs and arms would be a lot more versatile than a thing with caterpillar treads, except for the problem that nobody knows how to program the processor to be as smart as a human, or even an ape or dog. There is basically no common sense software, and no intelligence or learning capacity. These robots will never live up to their potential to serve us until they have human-like intelligence.
I think "more human" is a strange choice of words. It's a mistake to view robots in purely anthropomorphic terms.
On the other hand, I can certainly see how the iRobot is more of a real robot than the toylike Asimo.
Wow, the commenters here are mostly dead wrong.
If Asimo had been designed in the form of a platypus we would all be mocking the stupidity of the idea. Yet a platypus is no more likely to be the optimum design for a given task than a human is.
The guy is right. Constraining the design of a robot to make it look like a human is just plain stupid.
Really? You may want to check the validity of your statement before you insult everyone. While I agree that a notion of a robot should not be constrained to anthropomorphic design, there are many valid reason to pursue such robots to work in a human-inhabited world.
Also I wouldn't be sure if platypus robot would be ridiculed as you would suggest - robotics research has benefited much from biomimicry in the past, and will continue to do so. I think people will be impressed if a robot can do half the thing a platypus can do.
The main argument for a humanoid robot is that it can basically replace a human. Tools and appliances are generally designed to be used by a person; if you wanted a robot to use them instead, you'd have to design a specialized robot for each tool. Or, just make one humanoid robot that can use them all. It makes sense, and isn't stupid at all.
Some people say that the only way people will accept robots in everyday life will be if they are humanoid, so people can relate to them. But I think R2 D2, and even the Roomba, shows that you can make things shaped like trash cans relatable.
At the end of the day, this guy is basically saying that generalist robots are impractical, and that specialist robots are the way to go. For now, he is right, but its a short sighted view. ASIMO has improved quite a bit, its only a matter of time before it can perform as well as (and pobably better) than a person at everything it does.
I agree. Humanoid robots are good for fun, or for simply advancing technology, but when you're talking usefulness as it pertains to specific tasks, then a robot with a design targeted at performing said tasks, will inherently be better.
But what if you are talking a single robot as it pertains to general tasks in a human shaped world?
Sure Warrior can carry weight and survive drops. But can it fall off the stairs, crack its humanoid head open and go into comma, just like ASIMO?
This is essentially, "My robot can beat up your robot! Neener neener neener!"
Yup.
Actually, neither is human; they're both toasters.
How much for the iKill version? Is there a bulk discount?
Asimo is basically a Proof of Concept project that has yet to prove itself. Right now its sole purpose is to demonstrate how fantastic it is. Performing preplanned routines is well and good and, yes, impressive, but until you can point it to any set of stairs or any surface condition then it's not usefull.
All the commenters saying that a fully functional humanoid robot is infinitely more usefull are absolutely correct. However, we are decades away, at least, from a fully functional humanoid robot.
Also take not that iRobot's business IS robotics while Honda is automobiles. Even if you could buy an Asimo, you wouldn't be able to do anything with it.
...Because eleventy billion years of evolution can't be wrong.
Just like the space program yields real world tech, so does Asimo. Just take a look at an article on this very blog that shows human augmentation suits that will help the elderly, factory workers, and eventually amputees. And its based on Asimo tech. Asimo is a project to spawn very necessary tech.
On top of that, if you want a robot that just kills things, go with the one called warrior. I'm sure its very good at what it does. iRobot's blood thirstiness is the reason I'm boycotting their roombas. Asimov would be ashamed. On the other hand if you want a robot that will serve or work with humans you may very well want it to look human.
"Bloodthirstiness"? Creating robots to handle the more dangerous combat situations prevents blood from being shed, since a solider doesn't have to be put in harm's way. As for the whole Warrior vs. Asimo argument, I think what that CEO was trying to say was that his hunter-killer robot is better than a fancy two-legged Big Trak - which is exactly what Asimo is. Now, if Honda's little project was already in production doing the dishes or whatever, his argument would've been utterly bogus. But he was talking about a functional combat droid vs. a humanoid science project. My only real fear regarding the proliferation of robots in war would be temptation to solve all problems with a huge airdrop of tin soldiers. However, the incredible costs and logistic issues involved would probably keep things in check.
Two words for you dude: Mobile Dolls. Make war bloodless and you make it easier to wage. Something's got to keep us from invading countries over anything that bugs us.
With this logic why on earth would the automobile ever have been invented. A horse was just fine. At the turn of the 20th century why on earth would someone want a horseless carriage. It would hardly ever work and it would get stuck in the mud where a horse was more reliable and trustworthy.
I bet the guys in the horseless carriage business used to make transparent missives about Ford as well.
Traction-driven robotic devices will continue to be more useful in the short term, but Angle may want to exercise some caution on his dismissive approach to Japanese humanoids. Doesn’t he know that as the U.S. Department of Defense aims to develop its dream of robotic military supremacy, it also continues to cherish the dream of a robot soldier that looks, thinks and fights like a soldier? In fact, I can imagine a scenario in the not-too-distant future when the Pentagon will start taking serious leads from Japan’s top humanoid robotics research institutes, such as Honda and the National Institute of Advanced Industrial Science and Technology. There is little doubt that for all around utility, aesthetics, and cognitive abilities, the future belongs to the ASIMOs of the robotic world. See http://www.zygbotics.com/2009/04/20/irobot-vs-asimo/#more-576