Rossum's Universal Robots

After a brief mention of Karel Capek's 1920s play "R.U.R (Rossum's Universal Robots)" in Higher I.T. today I found myself reminiscing about good old Sci-Fi films and wondering about the future. Could, one day, we be buying intelligent robots to help clean the house, cook the food and help us better our lives in every possible way?

Rossum's Universal Robots centred on the idea of intelligent machines built to work in a factory on a desert island. The Robots are shown to portray emotions and human character traits and eventually rebel and take control of the factory. This theme of an oppressed minority rising up and taking power seems to be a prevalent theme in almost all Robot Based Science Fiction stories. In I, Robot (The film) Will Smith found himself facing off against hordes of robots which had been oppressed by being forced to adhere to the "3 Robot Commandments", when one of the robots "sunny" was built, the mainframe at the company which produced robots created sunny differently, so that he could disobey the commandments. Will Smith's character unveiled a massive conspiracy by the intelligent mainframe to give robots emotions and ensure that they did not have to obey the commandments. The film culminated in showers of pyro-technics and a dramatic scene at the top of the US Robots and Mechanical Men, Inc. building. The film differs greatly from the original collection of short stories by Isaac Asimov. Asimov's short stories originally focused on humans, machines and the moral implications of their interactions. Asimov questioned their rights, responsibilities and their liabilities. Could we one day be facing a society where man and machine interact, live and work together?

If machines were truly "intelligent", would they be based on human intelligence? If so, would they be designed with character traits? or would their programming allow them to develop character traits? I think that perhaps the latter would be the more likely option, because the human definition of "Intelligence" is:

"Main Entry: in·tel·li·gence
Pronunciation: in-'tel-&-j&n(t)s
Function: noun
1 a : the ability to learn or understand or to deal with new or trying situations
b : the ability to apply knowledge to manipulate one's environment or to think abstractly as measured by objective criteria (as tests)
2 : mental acuteness —in·tel·li·gent /in-'tel-&-j&nt/ adjective —in·tel·li·gent·ly adverb"

And so, one can assume that if an intellifent being can be deemed intelligent it gathers information for itself. How would a robot gather this information? we would need to give it senses, so that information can be input directly to the robot as opposed to a human engineer reprogramming it with the "knowledge". And so a robot (in order to be "intelligent") would have to have eyes, ears, nose and a sense of touch. In advanced models I guess a sense of taste would be included, but let's stick with the budget model.

So, myself not being the best student of the human body and workings thereof I cannot say that all of my ponderings will be entirely accurate. But in order for this intelligent robot to be able to learn it will need a massive amount of storage, like the human brain. Unlike the human brain we will not want the robot's storage to be able to be deleted or fade with age. The robot would need a processor preloaded with some sort of operating system. This processor would help categorise all of the information relayed from the senses into appropriate categories for later reference. The Operating System would be preloaded with innumerable categories and subcategories etc. so that the robot could store things accurately for later reference. This information in the storage would have to be readily accessible as well, so that in a time of need the robot would be able to use it swiftly. Which is why a second processor, running independantly of the first would have to be installed to swiftly access information and send it to the "brain" unit.

The "Brain" unit may be seen as the same as the storage area, but it is very different. The "Brain" unit would be the equivalent of an Algorhythm and Logic unit in your computer's CPU. The "Brain" would be the first receiver of the information from the senses, and it would also receive the information sent by the storage unit and "decide" by cross referencing with previous experiences which would be the best course of action to take in such a situation. This "brain" would probably have dual or triple processors running simultaneously to perform the calculations necessary for the robot to function. Up to now I have merely described a "thinking" and "intelligent" robot. But in true robot fashion, we'd need to have something that moves. And we'd need a power source for this robot. And, if we were the company producing the robot, we would want a deliberate flaw like longevity so that we could encourage as many sales as possible.

So first let's handle movement. Would we have a robot which was bipedular? i.e. another sci-fi manlike robot. Or would all the robobts be different according to their uses? I think that the best way forward would be for a standard model of robot, with specialised models available built to specification. The standard (and I mean bog-standard) model would use wheels, any number over 3 I imagine would be optimum. Otherwise stabilisers would come into play, making the cost rise. The wheeled models would encounter far too many problems in every day life, and would probably require stanna stair lifts to be installed throughout the country. So the uses of wheeled ones? Transport aids? Factory workers. Anything that doesn't involve stairs really.

The more popular model would probably be the humanoid, like C3-P0 here. The humanoid models could be used for as many tasks as humans can do, but would be more fragile. Humanoid bots would not be as commonplace as may be perceived however. In Science Fiction films they are the most popular design because the links between man and machine are more obvious, hence the director can portray the frailties of man more obviously. Star Wars robots would probably be the more likely models of robots running through our streets. A massive mish mash of robots all performing various tasks, all looking as individual as their owners. Humanoid bots would be used for companionship. I highly doubt that they would be used for interpretting like in star wars, due to the availability of hand-held translation devices, and so would be few and far between. But what of power supply? how do the robots run?

On sugar and spice and everything nice? No. What COULD they run on? So hypothetically, you have your robot, he's all set up and ready to go, how do you make him go? Well let's think, He (I'm not implying that the robot's not a girl, I'm just calling it a he because in my head it's C3-PO esque) will need oil to keep his limbs working. Well, that's if he's a 21st Century robot. I think that by the time that this could hypothetically happen, we could have a more exciting lubricant which can be used from renewable sources.

So this fluid of lubrication for joints etc. We will also need something to deliver electric jolts to everything to power them. Now, we can't fit him with a solar panel (advanced as these solar panels may be) he may spend great lengths of time out of the sun (consider the poor Scottish owned robot). So we need an internal power source which could run indefinitely? No, we want a power source which cuts out after X years so that people have to go out and buy the newest model. Don't forget boys and girls, whilst we're talking about the future, we're not talking about a true fantasy where people aren't out for profit. Don't kid yourself, if a form of perpetual energy was invented the consuming public would be the last to know. We need a power source that is internal, something that can last for X amount of time without fault, and something that is cheap. I think that perhaps we would be using some sort of battery, where you can just plug your robot into the wall to keep him going. Failing that there's always nuclear fusion.

But my favourite thought about keeping robots running, is a flight of fantasy about cyborgs. Just hear me through...

The human body works on electric charges pulsing around. The brain is firing constantly, and that crazy guy who jolted that frog's leg with electricity, remember? Well, what if, instead of oil and battery power, we gave this robot blood and a heart. Apparently all that tells the heart to keep pumping is the brain sending signals, hypothetically, the robot's "brain" could send the same signals. But this would require much rethinking about our robot. He would need lungs, and a stomach for intaking energy. Also, you would end up with Frankenstein's Monster. (which by the way is a really good read)

So what have we learned?? What do you need to build a robot?

1: A degree in computing
2: A degree in biochemistry
3: A time machine
4: Money
5: An army of volunteers
6: Tools, lots of tools
7: An Active imagination.
8: Wait 70/80 years and maybe you'll buy one.