AI is impressive: a humanoid robot learned to make coffee after watching a human do it (video)

On Sunday, January 7, Brett Adcock from Figure announced a “ChatGPT moment” for humanoid robots. Now, the robot can watch a person perform tasks, build its own understanding of how to perform them, and start doing them completely autonomously.

This was reported by New Atlas.

Universal humanoid robots will have to perform a variety of tasks. To do so, they will need to understand all the tools and devices, objects, methods, and goals that we humans use to get the job done.

Figure, like Tesla, Agility, and many other companies, is focused on self-sufficient, full-bodied humanoids that can theoretically walk into any workplace and eventually learn to do any human job.

On his page in the on social network X Adcock posted a video in which the Figure robot performs a task related to a coffee maker with a cup already in it. It responds to a spoken command, opens the top hatch, puts in a coffee pod, closes the hatch, presses a button, and allows the man who ordered the coffee to pull a full cup out of the machine himself.

“Figure-01 has learned how to make coffee. Our artificial intelligence learned this by watching people make coffee,” he said.

The Figure robot spent 10 hours learning the video, and now it can do it on its own. He added a new standalone action to his library that can be passed to any other Figure robot running on the same system.

Soon, the robot will be able to find a cup in the kitchen, check whether the coffee maker is on and whether there is enough water in it, brew coffee by pressing a button, and bring it to the table without spilling it.

The publication noted that if the Figure robot can truly observe and learn, humanity will experience a major boost to accelerate on the wild frontiers of commercial humanoid robotics.

Source espreso
You might also like
Comments
Loading...

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More