Nerd For Tech
Published in

Nerd For Tech

Google X Robots Start to Work

https://baike.baidu.com/item/%E6%82%9F%E7%A9%BA%E6%9C%BA%E5%99%A8%E4%BA%BA/22882822

Recently, Google X’s moon landing factory has developed a robot that can clean the tables in a cafe and monitor the meeting room! It seems that the era of everyday robots is approaching.

Actually, the cafe waiter robot is already used on the Google campus in Mountain View, California.

The robot came from Google X’s Everyday Robots project and was developed in the Moonshot Factory. The goal is to produce robots that serve people in daily life. Everyday they will send the robots to actual working scenes, such as coffee shops.

These robots look for tables in the cafe, place one of the wheels on the table, and take a few seconds to determine if anyone is sitting there until they find an empty table.

Next, the robot begins to wipe the table twice. The robot twists and unfolds its limbs, extends its arms to the table and covers the surface with a transparent disinfectant method. Then it retracts the arm and squeezes the redundant liquid into the bucket at the bottom. Once the task is completed, it starts to move and looks for the next empty table.

Since the robot has been officially on duty for several weeks, the cafe customers have become accustomed to the service.

Everyday Robots conundrum

In other areas, robots can navigate complex traffic routes, drive cars safer than humans, etc. However, it seems that in the world of Everyday Robots, it’s hard to go through a messy room and open a doorknob. It appears that the task is merely coffee table cleaning, but in fact, there are a series of operations inside. For example, when other people or objects block the path, the robot should find the correct solution as well.

For the robot, the question which needs to answer correctly is — is there enough space to move smoothly in this direction, or does it need to completely re-plan the route?

The American-Norwegian engineer Brondmo, who leads the entire Everyday Robots robotics team, believes it can be possible to create genuinely universal robots by making the latest advances in machine learning. Engineers use machine learning to recognize objects and then run simulations millions of times, which can compress weeks of testing into hours. This way can also help robots in the laboratory understand their actual working environment, and accumulate skills to solve the inevitable dilemmas of field operations.

For example, it takes less than 10 hours for door lock training, and once a robot learns something, the “knowledge” will be sent back to the entire collective cloud intelligence, and then all robots can use the same skill to lock the door.

Brondmo hopes that their robots can live and work with humans instead of the world of robots.

AI’s Reliance on High-quality Data

There is an important consensus in the AI industry:

The quality of the training data directly determines the performance of the final AI model. The more scalable and accurate data, the more robust algorithm will be.

With the acceleration of the commercialization of AI and the application of AI technologies such as assisted driving and customer service chatbot in all walks of life, the expectation of data quality in the special scenarios is getting higher and higher. High-quality labeled data would be one of the core competitiveness of AI companies.

If the general datasets used by the previous algorithm model are coarse grains, what the algorithm model needs at present is a customized nutritious meal. If companies want to further improve certain models’ commercialization, they must gradually move forward from the general dataset to create the unique one.

End

Outsource your data labeling tasks to ByteBridge, you can get the high-quality ML training datasets cheaper and faster!

  • Free Trial Without Credit Card: you can get your sample result in a fast turnaround, check the output, and give feedback directly to our project manager.
  • 100% Human Validated
  • Transparent & Standard Pricing: clear pricing is available(labor cost included)

Why not have a try?

Source:https://www.robot-china.com/news/202111/30/68642.html

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
ByteBridge

ByteBridge

Data labeling outsourced service: get your ML training datasets cheaper and faster!— https://bytebridge.io/#/