It is common to give human characteristics to in-animate objects. People name cars, boats, and even production robots. Sometimes, you get to the point where you associate behaviors of the object with free thinking, but in the end any robot or vehicle is a pile of metal that can only do what is instructed to do. Putting a face on a robot can make things easier for interaction, but is still up to the end user to give the correct direction. Personally, I always thought that R2 was creepier than C3PO.
I think the value of an anthropomorphic structure for a robot would be in it's ability to be functional over a wide variety of tasks. Thus by having one of these type robots, you save yourself a great deal of money and space by not needing multiple robots that were designed to have only one function.
Something you can program on the fly and that can adapt to most any situation.
But we've got a long way to go. Humans are pretty adept at so many things, robots will need to make huge improvements in dexterity, balance, etc. Not to mention all of this increased functionality will need to packaged neatly into a small, flexible, good looking, and lightweight (humanoid) housing.
The last paragraph of the article is the clincher. If the face is not perfectly human, then the interaction will fail.
Take the animated movie Polar Express. The animation was impressive, but when it came down to the faces of the characters, they tended to jar you out of the moment. There are some impressive face/head-only robots out there, but they still look mechanical. The face is astoundingly complex in its degrees of freedom. It's the feature humans learn almost immediately after birth to cue on.
I agree with with Alex. I'm not sure society is quite ready for the robot that looks too human or acts too human. It is still far too creepy. I also agree that function is key much more so than human-like form to making robots appealing for mainstream use. I, for one, would need to see proof that the robot would save me time, money, escape from mundane chores like laundry and dish washing before I would sink any money into one unless it was purely for the fun factor. I think those of us who aren't wowed by gadgetry are still highly skeptical that this weird looking contraptions can do the routine jobs better than a human set of hands.
well said !!! Forget the look and go for function.
I can care less if my vaccum looks like a disk or a mouse or a mop with LEDs. If its job to clean, I do not care how it looks.
Obviously if it's responsibility is to monitor childrens activity in the room and report on various changes, it should not look like a monster or a big black box. Then the look and ergonimics are important.
Generally its look should comply with the industry it is used in. Nowever the look is a secondary issue. Function FIRST.
I guess my question would be, "what functions are the humanoid robots designed to perform?".
Robot vacuum cleaners don't use the same vacuum that I do. If the robot is designed to funtion using the same tools and in essentially the same way as a human would , then we might find a humanoid shape and style; arms, legs, head etc useful. The robot cars tested by google do not have a robot sitting at the wheel. The robotic drones of the air force and support bots for the army don't use humanoid robots either. And the NASA robot just installed on the space station, what does that look like?
One might question exactly what a humanoid robot is better at that a purpose designed robot cannot do better.
I question the usefulness of any humanoid robot other than as a technology demonstrator. If the only reason is to make their interactions with people more friendly, then I think the designers are misjudging the ability of humans to work with and adapt to machines.
I think goal of humanoid robotic engineers is to tread lightly in the human looks department so as to avoid the creepy feeling that people will feel if they encounter an I-Robot type mechanical helper.
I think this whole form versus function debate re robotics is a false argument. Robots have to demonstrate functionality to succeed. Fixed-position factory robots have. "Humanoid" ones haven't thus far. Robotic floor-scraping vacuum cleaners, yes. "Walking" robots, not really (Asimo excepted, I guess). Quite frankly, robots which look too human give me the creeps.
Are they robots or androids? We're not exactly sure. Each talking, gesturing Geminoid looks exactly like a real individual, starting with their creator, professor Hiroshi Ishiguro of Osaka University in Japan.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.