SN

私人新聞日報Private News Daily

Private use only · Michael SoMichael So

本頁為 Michael So 私人自用,非公開發佈,亦非任何機構之產品。 This page is for the personal use of Michael So only. It is not a public release and is not a product of any organisation.

AI 科技AI & Tech

波士頓動力公布 Atlas 全新機械手,現代目標全球部署 25,000 部Boston Dynamics announces the new Atlas robotic arm, with modern plans to deploy 25,000 units worldwide

Korea JoongAng Daily;IEEE Spectrum ·2026-10-01

波士頓動力公布為 Atlas 設計嘅全新機械手,十三個自由度,重點放喺容易製造、耐用同成本控制,由原型轉量產後部分零件成本下跌六成至八成。機械人已喺美國喬治亞州車廠實地訓練,母公司計劃最終全球部署二萬五千部,美國廠房年產三萬部。Boston Dynamics announced a brand new robotic hand designed for Atlas, with thirteen degrees of freedom, focusing on ease of manufacturing, durability, and cost control. After transitioning from prototype to mass production, the cost of some parts dropped by 60% to 80%. The robot has already been trained on-site at an automobile plant in Georgia, USA, and the parent company plans to eventually deploy 25,000 units globally, with the U.S. factory producing 30,000 units annually.

Boston Dynamics’ humanoid robot Atlas are seen training to perform automotive manufacturing tasks in a video released on Sept. 30.SCREEN CAPTURE

Once known for flashy stunts, humanoid robot Atlas is now hard at work training for a very different role — building cars.

Boston Dynamics unveiled "Inside the Robotics Race," a documentary jointly produced with Hyundai Motor Group and Bloomberg Media Studios, on Wednesday.

In the clip, Atlas stops in front of a car and bends at the waist. It grips a wheel and tire with both hands, then hoists them into the air. At a workbench, it picks up components and moves them from one place to another. Around the vehicle, it works alongside another Atlas robot. Researchers standing nearby watch monitors and closely track the robots’ every movement.

The documentary was filmed at the Robotics Metaplant Application Center (RMAC) at Hyundai Motor Group Metaplant America in Ellabell, Georgia. The center, which began operations in September, trains Atlas to perform tasks required in automotive manufacturing and tests its capabilities before deploying the robot to actual production lines.

The RMAC shown in the video is set up to resemble a real manufacturing environment, with vehicle bodies, parts racks and workbenches. Atlas moves through the facility, repeatedly practicing how to pick up and transport objects. The footage also shows the robot handling heavy objects such as wheels and tires and multiple Atlas robots working in the same space.

Boston Dynamics chose automotive manufacturing as the first application for Atlas because the industry demands high levels of safety and reliability. After validating the robot’s performance in real-world manufacturing environments, the company intends to expand its use to other industries, including food and beverages and semiconductors.

AI plays a central role in helping Atlas master unfamiliar tasks. The robot combines its ability to maintain balance and control its arms and legs with the ability to understand instructions and determine what to do next. Rather than relying on the conventional approach of programming individual movements one by one, AI allows Atlas to learn a range of tasks and adapt to changes in its surroundings.

Collecting real-world data is another key role of the RMAC. Boston Dynamics said physical AI, much like large language models, becomes more capable as it gains access to more data, so real-world experience outside the lab is crucial for Atlas.

Efforts are also underway to bring down costs ahead of mass production. Boston Dynamics said it has reduced the cost of some components by 60 to 80 percent as it transitions Atlas from a prototype to a mass-produced product.

Hyundai Motor Group intends to build a facility in the United States capable of producing up to 30,000 robots annually and eventually deploy 25,000 Atlas robots across its operations worldwide.

BY HYEON YE-SEUL [kim.jiye@joongang.co.kr]

This article was originally written in Korean and translated by a bilingual reporter with the help of generative AI tools. It was then edited by a native English-speaking editor. All AI-assisted translations are reviewed and refined by our newsroom.

Boston Dynamics’ humanoid robot Atlas is seen training to perform automotive manufacturing tasks in a video released on Sept. 30. SCREEN CAPTURE

曾以炫目的特技而闻名的人形机器人Atlas,现在正努力训练,准备担任一个完全不同的角色——制造汽车。

Boston Dynamics unveiled "Inside the Robotics Race," a documentary jointly produced with Hyundai Motor Group and Bloomberg Media Studios, on Wednesday.

在片段中,Atlas 停在一辆车前,弯下腰。它用双手抓住一个轮子和轮胎,然后将它们举到空中。在工作台旁,它拿起零件并把它们从一个地方移到另一个地方。在车辆周围,它与另一台 Atlas 机器人一起工作。旁边的研究人员观看显示器,并仔细跟踪机器人的每一个动作。

The documentary was filmed at the Robotics Metaplant Application Center (RMAC) at Hyundai Motor Group Metaplant America in Ellabell, Georgia. The center, which began operations in September, trains Atlas to perform tasks required in automotive manufacturing and tests its capabilities before deploying the robot to actual production lines.

The RMAC shown in the video is set up to resemble a real manufacturing environment, with vehicle bodies, parts racks, and workbenches. Atlas moves through the facility, repeatedly practicing how to pick up and transport objects. The footage also shows the robot handling heavy objects such as wheels and tires and multiple Atlas robots working in the same space.

Boston Dynamics chose automotive manufacturing as the first application for Atlas because the industry requires high levels of safety and reliability. After verifying the robot’s performance in real-world manufacturing environments, the company plans to expand its use to other industries, including food and beverage and semiconductors.

AI plays a central role in helping Atlas master unfamiliar tasks. The robot combines its ability to maintain balance and control its arms and legs with the ability to understand instructions and determine what to do next. Rather than relying on the conventional approach of programming individual movements one by one, AI allows Atlas to learn a range of tasks and adapt to changes in its surroundings.

Collecting real-world data is another key role of the RMAC. Boston Dynamics said that physical AI, like large language models, becomes more capable as it gains access to more data, so real-world experience outside the lab is crucial for Atlas.

Efforts are also underway to reduce costs ahead of mass production. Boston Dynamics said it has cut the cost of some components by 60 to 80 percent as it transitions Atlas from a prototype to a mass-produced product.

Hyundai Motor Group plans to build a facility in the United States that can produce up to 30,000 robots annually and eventually deploy 25,000 Atlas robots across its global operations.

BY HYEON YE-SEUL [kim.jiye@joongang.co.kr]

This article was originally written in Korean and translated by a bilingual reporter with the help of generative AI tools. It was then edited by a native English-speaking editor. All AI-assisted translations are reviewed and refined by our newsroom.

原文出處:Source: Korea JoongAng Daily;IEEE Spectrum ↗