What do you get when you cross ingenuity with need? Cheap and effecitve solutions. The American soldier is well known for his (her) resourcefulness and using RC cars against IEDs is no exception:
A young private...has one of those radio-controlled toy cars. When they find unidentifiable debris in the road, E.S. sends out his little RC car and rams it. If it's light enough to be moved or knocked over, it's too light to be a bomb...
Ok, so now we have manually controlled, line of sight controlled vehicles (while the radio waves aren't line of sight, try controlling one of the cars around a corner or through a ditch 50m away). Add some remote sensing, or bettery yet autonomy and you have a cheap picket network of active sensors and investigators. This is nothing new and we've all seen this in the movie Minority Report when the silver spiders are looking for Tom Cruise. But those little guys swarmed and where not individuals (like the critters acting like Shatner's gremlins in Star Wars III, opening up the plane to destroy it).
The University of Pennsylvania received a $5m grant from DoD to develop swarming behavior in robots:
"Our objective here is to develop the software framework and tools for a new generation of autonomous robots, ultimately to the point where an operator can supervise an immense swarm of small robots through unfamiliar terrain," said Vijay Kumar, director of the GRASP Lab at Penn's School of Engineering and Applied Science and principal investigator of the Swarms Project.
Video of the robots working autonomously is interesting, but still, what of it? But does using energy efficient power resources, remote imaging, and inexpensive semi- or fully-autonomous vehicles to scout and establish a perimeter fit in modern warfare without frontlines? Soldiers carry inexpensive UAVs on the backs to launch recon missions, why not have these little devices monitoring w/ the ability to react?
More importantly is the logic model the truly autonomous vehicle will use. Should this vehicle, the RC car or the sentry robot respond to each potential threat in the same manner? Should it include some measurement in its calculus of determining action toward self-protection to protect another day?