X’s mission makes robots boring

The target of these creatures is the desktop. One of them will walk to a table and think for a few seconds to determine if anyone is seated; if so, it will keep going until it finds an empty one. After hovering for a second—perhaps algorithmically taking a deep breath before the “let’s do it” moment—the robot rotates and spreads its limbs, stretches its arms to the table, and methodically covers the surface with a transparent disinfectant . Then it retracted its arm and squeezed the excess liquid into the bucket on its base. When the task is completed, it moves on, looking for another table to swipe the card.

After lunch, people don’t even bother to look up. The robot has been doing this for several weeks.

Everyday Robots has built more than 100 robots at X’s Mountain View headquarters.

Photography: Michel Grosskopf

No, this is not a desperate attempt to solve the labor shortage.Its research Everyday robot, An item of X, Alphabet’s self-styled “Moon Landing Factory”. The cafe proving ground is one of Google’s dozens of campuses in Mountain View, California, and a small part of the company’s huge workforce has now returned to work. The project hopes to make robots useful and run in the field rather than in a controlled environment like a factory. After years of development, Everyday Robots finally sent its robot to the world-or at least walked out of the X headquarters building-to do actual work.Two years after Tom Simonite of WIRED, they invited me to observe that this is already a milestone Finally took a look at the projectAt that time, they asked the robot to sort the garbage into the appropriate recycling bins. The cleaning service represents the next, if not the last frontier.

Attack of the squeegee robot: X’s convoy wiped the tables in the Google Cafe.Video: Wired employees

Darcy Grinolds leads Everyday’s hardware reliability and design verification team.

Photography: Michel Grosskopf

I’m joking, but this is serious business. Everyday Robots is trying to do two very difficult things, a difficult challenge that some people question whether the effort is worthwhile. The first is to reliably perform the tasks of the human assistant.Everyday robots live on the edge of the razor Moravec Paradox, It pointed out that it is relatively easy for computers to perform difficult cognitive tasks, while it is very difficult to replicate the functions of a two-year-old child.In other places under the Alphabet umbrella, robots navigate complex traffic routes, safer than humans driving cars, and become Go championIn the world of Everyday Robots, conquering a mundane task, such as walking through a messy room and opening a tricky doorknob, is like winning the Super Bowl. For example, the activity of wiping a table is more than just sliding-it includes a whole set of operations that cause it. See what happens when the path is blocked by people or objects. “The correct response of the robot is, well, do I have enough space to move gracefully?” said Darcy Grinolds, who is the head of the hardware reliability and design verification team for the project. “Or do I need to change course completely?”

The second difficult thing the project is trying to do is to move towards this goal in such a way that, in terms of economy and efficiency, having a robot on hand is more meaningful than a boring and low-paid person.

Open the door to the future of robots.

Video: Wired employees

For more than a decade, Google and now X have been obsessed with pursuing this vision.Leading the Everyday Robots team is Hans Peter Brondmo, an engineer who grew up in Norway. He is an entrepreneur and engineer. He joined X in 2015 and had to understand how the former leader Andy Rubin acquired robotics. Discordant sound, the latter in Sexual Harassment Claim Cloud“Hans Peter is not the obvious choice,” said Astro Teller, CEO of X. “He cares about robotics, but he will be the first person to tell you that he is not a world-class robotics expert. I chose him because he is a world-class entrepreneur who really understands people. He is a complete socialist— -He is from Norway!”

In the office he shared with a non-functional robotic arm he built as a teenager, Brondmo explained that only with the latest advances in machine learning can it be possible to create an effective general-purpose robot. Engineers use machine learning to train software to recognize objects, then run millions of simulations, compressing weeks of testing into hours. This helps the clumsy robots in his laboratory truly understand their environment, and build a tool set based on this knowledge to help solve the inevitable dilemma in the wild. Although Everyday Robots may not be as gorgeous as the dystopian robots in the Boston Dynamics video, they are optimized to complete the task. (Alphabet once owned Boston Dynamics, but Sold 2017. )

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *