Atlas Can Work In A Factory Setting, New Video Reveals

BY FIGURE AI
Introducing Learned Natural Walking
We are excited to introduce our end-to-end neural network, trained with reinforcement learning (RL), for humanoid locomotion.

Atlas remains keen on proving why it’s the golden child of Boston Dynamics after achieving yet another impressive feat: independently maneuvering engine covers in a simulated factory setting.


While Boston Dynamics is already known for its slate of advanced robots, this latest Atlas development is a significant step towards the field of robot autonomy—especially now that Atlas has proven capable of handling repetitive tasks without the need for human supervision.


The new Atlas feat is documented in a new video, titled “Atlas Goes Hands On.” As appropriate to its pun, Atlas can now use its machine learning (ML) capabilities to locate and identify bins that contain engine covers. Atlas then uses its body to grasp the covers and even place them on a mobile dolly in a calculated and precise fashion.




Although seemingly an ordinary feat, the video showcases significant developments in the field of robotics:


  • Full Autonomy: Thanks to ML and real-time data, Atlas can now move independently without needing remote control. It can execute its actions based on data it receives.

  • Adaptability: Atlas can detect potential obstructions on its path and adapt to avoid any work mishaps. These include avoiding obstacles and shifting bins.

  • Precision Handling: Atlas pairs its advanced ML capabilities with precision grasping tools. The robot is dextrous enough to grasp and handle the engine covers without missing a spot.

Further advancements on Atlas has the potential to introduce groundbreaking strides in factory work. Thanks to robots like Atlas, workplaces are now room for more complex roles that require problem-solving and creativity - a role perfect for humans.

Atlas Can Work In A Factory Setting, New Video Reveals | Figure Robotics