When the Mars Phoenix Lander touched down on the Martian soil, discovered water ice and microbe-friendly acidic soil last month, it was quite a feat—for a human-controlled robot. But what if the robots sent to distant worlds could think for themselves, a la Wall-E and his Apple-esque main squeeze, EVE? NASA's betting on it, and has actually already started work on a "tier-scalable reconnaissance" program that would see armies of small probes exploring the outer reaches of our solar system with minimal human intervention, if any at all. Of course, such self-sufficient robots wouldn't be as cute as Wall-E—they'd actually be large hivemind dirigibles controlling an army of autonomous planet-side probes (think: Rush Limbaugh's radio program)—but nevertheless, the possibility for intergalactic robot love stories has never been closer to reality.

The experimental tier-scalable reconnaissance process begins with an orbiting spacecraft deciding on its own where to send an airship armed with further probes. The airship itself, once deployed planetside, could also override the orbiter and decide on its own where to send its stable of landers, NASA said. The first real-world example of the process is set to blast off in 2009. Called the Sky Crane carrier, the probe will hover above the surface of Mars (it's so hot right now!) on retrorockets before lowering an "SUV-sized rover using a winch and tether" onto the ideal landing site.

Meanwhile, back here on earth scientists at Caltech have started testing new pseudo self-aware algorithms with three small rovers and a camera that surveys a simulated indoor landscape. The camera identifies targets of opportunity and obstacles, and in turn commands the rovers to drive around obstacles to reach targets without an ounce of human intervention. Eventually, the camera will be replaced with the hivemind airship. The cylon wars will follow soon afterward, we imagine. [MSNBC]