Already touting a 1-million-robot arsenal across its fulfillment network, Amazon is leaning into more robotics tech aimed at hastening warehouse tasks and facilitating faster deliveries.
In multiple blog posts on Wednesday, Amazon introduced the Blue Jay robotics system that coordinates numerous arms to perform picking, stowing and task consolidation, which the company says “effectively collapses three assembly lines into one.”
More from Sourcing Journal
Blue Jay is already being tested in production at an unidentified Amazon fulfillment center in South Carolina, where it’s already able to pick, stow and consolidate approximately 75 percent of items stored at the site.
Amazon expects the system to serve as a core technology to help power the company’s same-day delivery sites . Overall, the company has 595 package delivery stations already active as of the first quarter, alongside 65 sub-same-day fulfillment centers , according to estimates from supply chain consulting firm MWPVL International.
According to the e-commerce giant, the technology will allow employees to shift from repetitive physical tasks like stowing items to “higher-value” work like quality control and problem-solving—all while making jobs less physically demanding by reducing repetitive reaching and lifting.
“Visually, Blue Jay operates like a juggler who never drops a ball—only here, the ‘balls’ are tens of thousands of items moving at high speed,” said the blog post, which indicated that the system’s development moved from concept to product in just over a year.
The company credits AI advancements and the usage of digital twins in hastening the development process, which usually took three or more years for Amazon robotics systems like Cardinal , Sparrow or Robin. Those robotics solutions all use computer vision and suction cups to move individual products or packages packed by human workers.
The blog posts were released a day after the New York Times reported that the tech titan’s extensive automation plans could save the company from having to hire 160,000 employees by 2027. When looking further out, Amazon could skip out on hiring 600,000 workers by 2023, according to the Times, which reported on internal documents from the Big Tech firm.
That report indicated that Amazon would save roughly 30 cents per item it packs, picks and delivers to customers. Amazon told Sourcing Journal that the documents were misleading and incomplete, and that they “reflect the perspective of just one team and don’t represent our overall hiring strategy.”
Whether the hiring projections come to fruition or not, Morgan Stanley estimates that cost savings from the automation ambitions could result in between $2 billion and $4 billion in cost savings if 40 delivery fulfillment centers were equipped with next-gen robotics by the end of 2026.
Blue Jay follows other recent robotics developments at Amazon , including the testing the humanoid robot Digit and the inventory processing Sequoia robotics system.
In May, it debuted Vulcan, a robotic system that has a sense of touch, and shortly after rolled out DeepFleet, an AI model that coordinates large fleets of mobile robots with different form factors across facilities. There was also a report over the summer that Amazon was planning a “humanoid park” in a San Francisco office to train humanoid robots via AI to deliver parcels to consumers’ homes.
Beyond the Blue Jay announcement, Amazon also unveiled a new agentic AI model designed to help fulfillment operations teams make better decisions across facilities.
The model, called Project Eluna, is built to help operators anticipate supply chain bottlenecks by pulling in historical and real-time data to recommend actions, ideally cutting down time needed to analyze dashboards so users can spend more time coaching their teams.
Project Eluna will be piloted at an undisclosed fulfillment center in Tennessee this holiday season, and will initially assist operators with sortation optimization. According to the blog post, operators can ask questions like, “Where should we shift people to avoid a bottleneck?”
Amazon says the agentic AI model will also eventually support preventive safety measures, helping plan ergonomic employee rotations and improve maintenance schedules.
Amazon confirms trial of smart glasses for delivery drivers
And on the delivery front, the e-commerce giant is developing smart glasses for its drivers.
The smart glasses, which Amazon has not named, leverage AI-powered sensing capabilities and computer vision, along with cameras to create a heads-up display that shows essential information directly in the driver’s field of vision.
This can enable them to scan packages, follow turn-by-turn walking directions and capture proof of delivery all without needing to use a smartphone.
Hundreds of Amazon delivery drivers tested early versions of the glasses and provided feedback to shape the technology, whether it be the clarity of the displays or the comfort of the rims.
The glasses feature a small controller worn in the delivery vest that contains operational controls, a swappable battery ensuring all-day use, and a dedicated emergency button to reach emergency services along their routes if needed. The glasses also support prescription lenses along with transitional lenses that automatically adjust to light.
Amazon is currently trialing the glasses with delivery drivers in North America and plans to refine the technology before a wider rollout.
The company has not indicated when the smart glasses will be deployed at mass scale. Reuters first reported last November that the glasses were in development.
Amazon anticipates future versions of the glasses will provide real-time defect detection, where the glasses can help notify drivers if they’ve mistakenly dropped a package at a customer doorstep that does not correspond with the house or apartment number on the package. They are also anticipated to detect hazards like low light so lenses can adjust on the fly, and notify the users if there’s a pet in the yard.
