Despite the US Department of Defense once predicting that a third of US fighting strength would be composed of robots by 2015, experts warn that the machine wars seen in the movies will remain science fiction for quite some time yet.
"I'll be back" said Arnold Schwarzenegger as cyborg-assassin the Terminator, back from the year 2029 to carry out a murder in 1984. But it seems that, when it comes to science fact rather than science fiction, it is unlikely that anything like him - or should he be an it? - will ever "be" at all.
Robots in the home have been promised for a while and though - as the BBC's Jon Stewart has discovered - technology is slowly allowing robots closer to domestic use, some of the most practical applications so far have been in military operations.
Robots so far are not quite up to the images depicted in science fiction
What robots are doing in modern warfare is no small feat. Machines undertake bomb disposal, mine detection and entering unknown places of interest before sending in soldiers - a practice that the military believes is saving lives.
Unmanned Aerial Vehicles (UAVs) or drones have already carried out military operations carrying "lethal payloads" without having a soldier on board. It was called "the only game in town in terms of confronting or trying to disrupt the al Qaeda leadership [in Pakistan]" by CIA director Leon Panetta in 2009.
Despite it being called a "drone war" by some commentators, the Department of Defense is keen to point towards its key goals of "intelligence, reconnaissance and surveillance" when it is using robotics.
In a document setting out an unmanned military vehicle roadmap between 2005 and 2030, it stated that around a third of US fighting strength would be composed of robots in a $127bn (£80bn) project.
This was scrapped in 2009.
But some of the latest prototypes could still make robots in military operations on the ground a regular occurrence.
The LS3 - known as "Alphadog" to its developers - will be able to carry up to 400lbs (180kg) of equipment over a distance of up to 20 miles (32km) over a 24-hour period.
In practical terms, this means that it could replace the use of a mule or donkey to carry heavy loads.
Newest prototypes move a lot more like organic animals than previously
Funded by the Defense Advanced Research Projects Agency (Darpa) and the US Marine Corps, it can withstand being kicked and maintains its balance when pushed.
"For years, I did work on one-legged hopping machines - funny contraptions in the lab," Marc Raibert, of Boston Dynamics told an audience at Stanford University.
"The number of remaining problems that need to be solved is a small but manageable list. In the next couple of years, I can imagine getting legged robot technology… out there and into use."
The way it is able to right itself when toppled and the noise it makes when moving is more than a little like the Terminator and many reacting to the clips have been amazed at just how realistically the robot moves.
Tech site Gizmodo called it the "creepiest and most awesome quadruped robot of all time".
Life and death
But despite the occasional YouTube comment saying this development will lead to the end of the world, this is where robotics experts are keen to see the comparisons with sci-fi stop.
"With today's technology, there has to a person in the loop with regard to a decision that would have the gravity of life and death," says Joe Dyer, chief operating officer of robotics firm iRobot.
"Will we ever have machines that would truly challenge Asimov's laws? Maybe, but it's going to be a long time coming."
The real stumbling block for robotics engineers is that where a robot fails - and a human excels - is context and recognition. People can tell things apart quickly, effectively and from a very early age.
While handling rough terrain is now generally considered to be at the later stages of development for robots, actual intelligence and recognition about what it finds when it reaches its destination is a lot more complicated.
"We're starting to nibble at the edges of it," says Dyer.
"The world will begin to change drastically when robots have the manipulation capability of a five or six-year-old child. That means you can start to do the basics."
Brain power
At the moment, "stupid" robots serve a purpose. Their main function is to go into places where it would be dangerous or impossible for a human to tread. High intelligence is not required so much as brute strength and the ability to keep humans away from potentially harmful situations.
Some robots are a lot less frightening to people than others
The first major mass practical use for these type of robots came in the aftermath of 9/11 as a search and rescue tool when emergency personnel were unable to carry out an operation. They were operated by human controllers in a similar way to how they are today.
But if futurist and author Ray Kurzweil is right, by 2019 a $1,000 (£650) computer will at least match the processing power of the human brain.
And this could lead to "intelligent" robots with an autonomy that many find uncomfortable. But what many see this as frightening at the moment, some experts think that it is only a matter of time before it becomes an idea that people will get used to and eventually consider normal.
"We [humans] don't like to give up our special-ness," robotics pioneer Rodney Brooks said in a Ted talk.
"Having the idea that robots could really have emotions, or that robots could be living creatures - I think is going to be hard for us to accept. But we're going to come to accept it over the next 50 years or so."
View the original article here
