Podcast: Play in new window
BOB HIRSHON (host):
Self-aware robots…I’m Bob Hirshon and this is Science Update.
We generally think of robots as automatons, programmed to do the bidding of their human creators. But roboticist Hod Lipson of Cornell University says that a new generation of machines can learn from its mistakes and even become self-aware.
HOD LIPSON (Cornell University):
What we are trying to do is to make machines that can learn from the environment. Let the machines adapt the way they think or the way they learn over time, based on their experience of what works and what doesn’t work. One project is about a machine that moves a leg for example and senses the acceleration that it feels, and over time, it begins to form an internal picture if you like of itself, where its legs are, how they work and so forth. It helps the robot makes decisions what it can and can’t do.
HIRSHON:
He says the goal is to eventually create machines that are more reliable and can adapt to unexpected situations. I’m Bob Hirshon for AAAS, the Science Society.