- Amazon’s robots handled 224 units/hours compared to people on 243 units/hour
- Robots are more consistent but people can sometimes be faster
- Further research is needed to prevent damage
After a series of robotic messages in the last few years, including its latest Vulcan robot, which has a sense of touch, Tester Amazon is testing the next generation models called ‘Stow’ and ‘Pick’, and their names say it all-they are designed to help with stew and picking processes for fulfillment centers.
However, the company has acknowledged that the current robotics technology has no place to replace human workers despite the fact that they achieved promising results.
According to its performance testing, humans were an average of 243 units per hour compared to robots with 224 units per hour. The difference is small, but Amazon also highlighted some nuances.
Amazon’s robots are good but not human good
Where the robots stand out are in consistency. “It was also found that humans had greater variation in stewing speeds: People can quickly stow many small objects effectively, but are slower with large objects, creeping after lower trash cans, or when using a step -free to reach the top trash,” Amazon explained.
The company also observed how people can perform multiple tasks at once with two hands, such as pulling out a storage tray with one hand and stowing a product with the other – something that the current generation of robots is unable to do.
Still closes the gorge, with Amazon’s robots stowing with more than 85% success with a similar stuch speed as humans in more than half a million tests.
That said, there are some challenges with the implementation of robotics across Amazon stocks, including damage caused by dropping products, inserting products into stew trolleys and crushing light boxes.
As such, although full human replacement is not yet possible, hybrid systems turn out to be very effective, with more work needed to improve handling and reduce damage caused by robots.
“Deeper research into predicting complex object interaction may be required in scaling manipulation more general,” Amazon’s researchers added.
