The American e-commerce giant processes an incessant flow of items every day in its various warehouses located around the world. To go faster, Amazon is currently relying on the barcode technology used by its robotic arms to process the various products to be delivered.
However, this option is far from offering optimal results. This situation has therefore led Amazon to consider other solutions that are better suited to the characteristics of its sorting robots. In a press release issued on Friday, Amazon not only recording bad news indicated that he had found another technology that should better suit his expectations. It would be an algorithm overtrained using AI, capable of optimizing the performance of its robots.
An AI-based system to make robotic arms more efficient
The e-commerce giant claims to have a computer system capable of studying items that pass one by one on conveyor belts. According to information obtained this Friday, the computer tool developed with the help of an expert in artificial intelligence (AI) will above all make it possible to solve the current deficit of robots.
Indeed, the latter are still not able to correctly read the various barcodes affixed to the articles in order to sort them correctly. This leads to errors in sending items to customers. Amazon’s goal is therefore to completely dispense with the use of barcodes in order to optimize its control chains. Nontas Antonakos, Applied Science Manager at Amazon’s Computer Vision Group in Berlin, said:
“Solving this problem, so that robots can pick up items and process them without needing to find and scan a barcode, is fundamental. This will help us get packages to customers faster and more accurately.”
Obviously, the solution to this problem takes the form of a system called “multimodal identification”. To get there, Amazon’s AI experts had to start by building a library of product images. The images as well as the data relating to the dimensions of the products fed the first versions of the algorithm. Subsequently, the use of cameras to continuously capture new images of items served to train the model.
A test phase and a 99% accuracy rate according to Amazon
When first used, Amazon’s estimates of the algorithm’s accuracy rate were between 75 and 80 percent. Beginnings that the company had then considered “promising”. In its press release, Amazon claims that the system’s accuracy rate now peaks at 99%. Indeed, since the first hitch that occurred during a Prime Day promotion, the algorithm has greatly improved. Thanks to some modifications made by the development team, the vision identification system can assign confidence ratings to its evaluations.
However, all is not perfect. According to Amazon’s AI team, it will be difficult to fine-tune the identification system to assess products handled by people. This is why the final objective is to have them manipulated directly by robots.
Before rolling out to all Amazon businesses globally, the company is putting the computer vision algorithm through a testing phase at its facilities in Barcelona and Hamburg. The first benefits are already being felt. Indeed, Amazon claims that the multimodal identification system has helped to speed up the processing time for packages in these two cities. Once it is sufficiently run-in, we imagine that the technology will be shared across all of the company’s facilities.
Source : CNET
We would love to thank the writer of this short article for this amazing material
Amazon is working on a technology that will replace barcodes
Our social media pages here and additional related pages here.https://www.ai-magazine.com/related-pages/