According to Wikipedia (http://en.wikipedia.org/wiki/Nanite), the word "nanite" is just one of several words used to describe very small machines. Research is underway to create such machines, continually making operating devices smaller and smaller.
Of course, as of now, nanites are nowhere near as small as they can be in theory. In theory and in the realm of science fiction, a fully-functioning machine could be so small as to be tinier than the smallest bacterium. And these tiny machines could be designed to reproduce themselves automatically, which over time would allow enough of them to form to accomplish almost any purpose.
A machine that small would be an ideal candidate for interstellar space travel. Why?
That's because the most straightforward approach to travel to other stars is to simply go through space as fast as possible, the closer to the speed of light, the better. Light in a vacuum moves at around 186,000 miles per second, roughly 6 trillion miles a year. Even so the closest star system is over four years away at such a speed. So any velocity significantly short of light speed involves waiting a long time to get anywhere.
To science fiction fans, there's nothing new in what I just said. But by means of numerous stories featuring starships routinely cruising up to light speed, fans may have been lulled into thinking that getting a spaceship up to the speed of light is relatively easy, given the right energy source. It actually isn't.
A realistic design to get a spacecraft up to 92% the speed of light, the theoretical Project Valkyrie (http://en.wikipedia.org/wiki/Project_Valkyrie), would involve a craft weighing only 100 tons, far smaller than the starship Enterprise. Even so, the estimated amount of antimatter required to get this craft up to speed is another 100 tons! So half the weight of the craft would be antimatter fuel...and antimatter happens to be the single most expensive substance humans can manufacture (http://en.wikipedia.org/wiki/Antimatter#Cost). Making 100 tons of it is completely beyond even the most outlandish of human plans (science fiction writers get to skip this messy detail and simply assume we've already got a pile of it).
On the other hand, a nanite, being so small, is much easier to get up to light speed. In fact we've got particle accelerators right now that move ions of heavy elements like gold up to speeds approaching that of light (http://en.wikipedia.org/wiki/Relativistic_Heavy_Ion_Collider).
So we can imagine perhaps in the not-too distant future, someone could assemble a particle accelerator designed to hurl nanites out to the unknown reaches of interstellar space fast enough to arrive at other stars in a matter of years instead of centuries. Once there, the nanites in theory could reproduce from elements they find wherever they land and eventually assemble themselves into a larger machine, one that could perhaps beam back pictures to us of the new worlds they're exploring.
Here's a problem with that thought--isn't it true, historically speaking, that harnessing forces to destroy is easier than wending the forces of nature to build? It's easier use fuel and air to make a bomb than it is to make a jet engine; it's easier to split the atom to destroy Hiroshima than it is to build an atomic power plant.
If we could accelerate nanites to other worlds to build probes, it would be even easier to send them on a mission to destroy, to consume all life they encounter, to seek out compounds from living bodies to reproduce swarms of new nanites. This idea has the potential to change a lot of science fiction. Having trouble with the Klingons? Simply fire some flesh-eating nanites at their home world...
Independence Day, the movie that features massive alien spaceships cruising through our atmosphere and blasting us with energy beams, would have it all wrong. The aliens would instead land on the planet with tiny particles too small for us to detect. These nano invaders would reproduce and reproduce, eating life on a small scale at first. Before they were even visible to the naked eye there would be billions of them, able to continue killing if even one of them survived any attempt to destroy them with radiation or chemicals. Our military would be useless, our medicines inadequate. Before long, all of mankind would be enveloped in a continually self-replicating tide of tiny devouring machines...
And the interesting thing about it is--this possible nanite space attack would not only be much more deadly than alien attacks as usually portrayed in science fiction, from a strictly scientific point of view it would also be much more likely...