Dortmund, at a program in GermanyToday, Amazon announced a new robotic system called Vulcan, which the company is “calling its first robot system with a real sense of touch – how to change how the robots interact with the physical world.” For the moderate period, the physical world that is most concerned with Amazon is with its warehouses, and the vulcan is designed to assist (or to take) the items in their mobile robotic inventory systems.
In two upcoming letters IEEE transaction on roboticsAmazon researchers describe how both sting and picking sides of the system operate. We covered Stowing in detail a few years ago, when we talked Aaron RadnessDirector of Applied Science in Amazon Robotics. Perns and his team have made great progress on stopping since then, at this point the operation warehouses have improved speed and reliability in over 500,000 stoos, where the average stowing robot is now slightly faster than the average sting human. W.E spoke with Parnaness to get an update on stoying.
Amazon’s stating process adaptation
Stowing is the process by which Amazon brings products to its warehouses and adds them to its inventory so that you can order them. Not surprisingly, Amazon has gone to extreme length to customize this process to maximize efficiency in both space and time. Human stores are presented with one Mobile robot pod The fabric is filled with cubis (compartment), which consists of elastic bands in front of them to keep the goods from falling out. Human’s job is to find a promising place in a bin, draws the plastic band aside, and makes the cheese in that place. The new house of the item is recorded in Amazon’s system, the pod then goes back to the warehouse, and the next pod comes with, ready for the next item.
Various manipulations are used to interact with human-oriented compartment.Heroic
The new paper on the sting includes some interesting numbers about Amazon’s inventory handling process that help keep the problem scale in perspective. More than 14 billion items are handed by hand at Amazon Warehouse every year. Amazon is hoping that the vulcan robots will be able to stove 80 percent of these items per day at the rate of 300 items, while operating 20 hours per day. This is a very, very high bar.
After a lot of practice, Amazon’s robots are now quite good in the sting task. Butness tells us that the Sto System is working three times faster because it was 18 months ago, meaning it is actually a little bit. And sharp An average compared to humans. It is exciting, but as Paranees states, expert humans still keep the robot in shame. “The fastest person in this task is like the Olympic athletes. They are much faster than the robot, and they are capable of storing the item in the pods at much density.” High density is important because it means more accessories can fit into warehouses that are physically close to people, especially relevant in urban areas where space is at a premium. The best people can be very creative when it comes to this physical three-dimensional “Tetris“The robots are still working.
https://www.youtube.com/watch?v=owxco05EK28
Where robots do Excels are planning ahead, and it is likely that the average robot stower is now able to beat the average human stower-tetris-NG is also a mental process. In the same way that good tetric players are thinking about where next The piece is about to go, not only the current piece, robots are capable of taking advantage of much information, which can optimize more information than humans what and when, where and when, what happens, says Parnena. “When you are a person doing this task, you get a buffer of 20 or 30 objects, and you are looking for an opportunity to fit those items into separate compartments, and to remember which place can go to the place. But the robot knows all the qualities of all our objects at once, and we can see all the bins with all the bins.”
Essentially, robots are far better in adaptation within the planning side of tetriaging, while humans are better in the manipulation side (still), but this difference is shutting down because robots become more experienced in dislocation and operation in operation. Amazon has worked to collect training data at Live Warehouse in the state of Germany and Washington for more than a year, and those robots have successfully chosen hundreds of thousands of items.
From your site articles
Related articles around web
