Boston Dynamics has shared a new video showing how its updated Atlas robot is able to competently handle workplace tasks using AI and machine learning tools that include reinforcement learning and computer vision.
The footage shows Atlas operating autonomously, moving engine covers between containers in a relatively fast and efficient manner.
“Atlas uses a machine learning (ML) vision model to detect and localize the environment fixtures and individual bins,” Boston Dynamics said in notes accompanying the video. “The robot uses a specialized grasping policy and continuously estimates the state of manipulated objects to achieve the task.”
The Hyundai-owned company emphasizes that there are no prescribed or teleoperated movements, and that all of the motions are generated autonomously online.
The video also shows how the bipedal humanoid robot is able to recover from so-called “action failures” — for example, when a task fails at the first attempt or if it trips on something — by using a combination of vision, force, and sensors.
The roboticist team at the Massachusetts-based company retired the hydraulic-powered Atlas in April and immediately replaced it with this fully electric version that it said is stronger, more dexterous, and more agile, and “able to move in ways that exceed human capabilities.”
Boston Dynamics has said that it wants to commercialize the robot in a similar way to how it’s deployed its dog-like Spot robot in industrial settings.
“This journey will start with Hyundai,” Boston Dynamics said in the spring. “In addition to investing in us, the Hyundai team is building the next generation of automotive manufacturing capabilities, and it will serve as a perfect testing ground for new Atlas applications. In the months and years ahead, we’re excited to show what the world’s most dynamic humanoid robot can really do — in the lab, in the factory, and in our lives.”
While human workers may watch Atlas with a degree of nervousness, companies deploying advanced robotic technology often claim that it’ll be used alongside human workers, or for mundane tasks so that human employees can focus on more meaningful duties. Only time will tell if those claims hold true.