Hundreds of robotic spheres that could one day explore planets like Mars are to be tested out by scientists back on Earth.
The micro-robots could land on the surface of another planet arranged in a capsule like eggs in a carton. Or they could be dropped onto the planet by a balloon floating above the surface. They would move by rolling and bouncing, powered by artificial muscles that alter their overall shape.
Although such a planetary mission may not occur for another 10 years at least, scientists are nearly ready to test the concept. With funding from the NASA Institute for Advanced Concepts, Steven Dubowsky of MIT in the US, and colleagues are developing prototype micro-bots. Lab tests are scheduled for September 2006 and, a year after that, a set of caves in New Mexico will offer a more challenging environment.
"The reason they're very attractive for the caves is a normal rover wouldn't be able to go down there and certainly wouldn't be able to come back up," Dubowsky told New Scientist. The prototypes are weighted on one side so that they always come to rest in a particular orientation, like Weebles. This leaves them sitting on top of a device that rapidly extends to make the robot jump.
In cave exploration, the micro-robots could position themselves to enable communications from deep within a cave. Each one would act as a relay, passing messages back to a central unit. Fifty micro-robots lined up in this manner could theoretically explore a cavern a kilometre deep.
The final bots would each be a few centimetres in diameter, about 100 grams in weight, and would be able to jump about 1.5 metres in a single bound, travelling 50 kilometres over their lifetimes.
Each robot would come equipped with a suite of scientific instruments, such as cameras and spectrometers, and would be able to communicate with one another, their designers claim.
The spherical micro-robots should be able to tackle the sort of terrain that is impossible for today's planetary rovers. They could also perform their research collectively, meaning the loss of a few bots would not jeopardise the overall mission.
Today's planetary rovers must find a flat region for landing, to avoid tipping over, and the loss of a single rover can cost a space agency hundreds of millions of dollars.
Since the micro-robots would communicate with one another and potentially determine their own next objective, the role of ground control crews would change, Dubowsky predicts. Existing methods are "very expensive and time consuming", he says.
All comments should respect the New Scientist House Rules. If you think a particular comment breaks these rules then please use the "Report" link in that comment to report it to us.
If you are having a technical problem posting a comment, please contact technical support.
