Researchers are experimenting with robots to help speed up the restoration of coral reefs.

Their researchers have been training an artificial intelligence to control collaborative robots (cobots), which work closely alongside humans.

“Some of these processes in coral propagation are just repetitive pick and place tasks, and they’re ideally suited to robotic automation,” says Ms Foster

A robotic arm can graft or glue coral fragments to the seed plugs. Another places them in the base, using vision systems to make decisions about how to grab it.

“Every piece of coral is different, even within the same species, so the robots need to recognise coral fragments and how to handle them,” says Nic Carey, senior principal research scientist at Autodesk.

“So far, they’re very good at handling the variability in coral shapes.”

  • treefrog
    link
    fedilink
    English
    arrow-up
    8
    ·
    10 months ago

    There’s a difference. When you program a machine it follows rigid logic. It’s predictable.

    When you train a machine it does not. It can make its own inferences and operate outside of strict parameters. It can also make bad inferences, what we call AI hallucinations.

    I don’t know that what you’re saying is wrong about avoiding responsibility, but programmed is not the right word for what’s basically a genie in a bottle. And we still hold accountable the member of the tribe that lets the genie out, or should anyway.