A research team led by University of Chicago geoscientist Patrick McGuire has successfully tested a feature-identifying system that could one day be used by "cyborg astrobiologists." The algorithms were able to pick out lichen from surrounding rock, and could be capable of handling other types of data. The team has incorporated a Hopfield neural network, a type of artificial intelligence for finding patterns in incoming data, into the system.
McGuire envisions space explorers wearing data-crunching Hopfield networks on their hips. "You would have a very complex artificial intelligence system, with access to different remote sensing databases, to field work that's been done before in the area, and it would have the ability to reason about these in human-like ways," he says.
McGuire hopes to train the network to process different textures, and then conduct analysis ranging from the microscopic to landscape-wide scales. The system could be used to search for Martian meteorites on Earth or uploaded to Mars-roving robots, until humans are ready to explore the surface of the planet on their own.
From Wired News
View Full Article
Abstracts Copyright © 2009 Information Inc., Bethesda, Maryland, USA